VISUOMOTOR FUNCTIONAL CONNECTIVITY: A TMS STUDY
Strigaro G1,2, Ruge D1, Chen JC1, Marshall L1, Desikan M1, Cantello R2, Rothwell JC1
1. Sobell Department of Motor Neuroscience and Movement Disorders, University College London Institute of Neurology, London, United Kingdom
2. Department of Translational Medicine, Section of Neurology, University of Piemonte Orientale ‘‘A. Avogadro’’, Novara, Italy
Corresponding author: Gionata Strigaro, MD
Department of Translational Medicine, Section of Neurology University of Piemonte Orientale “A. Avogadro”
Via Solaroli 17, 28100 Novara, Italy E-mail: [email protected]
Running title: Visuomotor functional connectivity
Key words: visual cortex; motor cortex; functional connectivity; transcranial magnetic stimulation. Word Count:
Key points
We studied the functional connectivity between primary visual and primary motor cortex using paired transcranial magnetic stimulation (TMS).
The connection is inhibitory at rest and possibly mediated by inhibitory interneurones in motor cortex. The effect reverses into facilitation during a visuomotor, but not audiomotor reaction task.
We conclude that a physiologically relevant occipitomotor connection can be activated by means of TMS. It may contribute to visuomotor integration as well as being involved in certain types of visual epilepsy.
Abstract
The interaction between the visual and the motor system is of crucial importance in motor control. To shed some light on its neural bases, we studied the functional connectivity between primary visual and primary motor cortex using paired transcranial magnetic stimulation (TMS).
Sixteen healthy volunteers (7 women, 21–51 yrs) participated in this study. Motor evoked potentials (MEPs) were recorded from the right first dorsal interosseous (FDI). We delivered a conditioning stimulus (CS) over the phosphene hot spot of the visual cortex, followed by a test shock (TS) over the left hand motor area at random interstimulus intervals (ISIs, 3 to 40 ms) while participants eyes were open or closed. A control study was also performed with the CS slightly lateral to Pz. The effects of the paired (CS+TS) stimulation were re-tested during visual and auditory reaction time tasks. Finally, we measured the effects of CS on short-interval intracortical inhibition (SICI) in the left M1.
At rest, independent of eye state, a visual CS significantly (p < 0.001) suppressed test MEPs, at ISIs from 18 to 40 ms. No effect was seen after conditioning lateral to Pz. In the visual, but not the auditory, reaction-time task, inhibition was replaced by a time-specific facilitation (p < 0.001). In addition, a visual CS increased SICI in left M1 at an ISI of 40 ms (p < 0.01), but not at 18 ms.
We conclude that it is possible to study a functional connection from visual to motor cortices using paired TMS. The connection is inhibitory at rest and possibly mediated by inhibitory interneurones in motor cortex. The effect reverses into facilitation during a visuomotor, but
not audiomotor reaction task. This suggests that it plays a role in visuomotor integration, and underlies some of the pathological connectivity previously described in neurological disease. Abbreviations
ADM, abductor digiti minimi; AMT, active motor threshold; ANOVA, analysis of variance; APB, abductor pollicis brevis; CS, conditioning stimulus; EMG, electromyography; FDI, first dorsal interosseous; ISI, interstimulus interval; LTD, long term depression; LTP, long term potentiation; MEP, motor evoked potential; rmANOVA, repeated measures of ANOVA; RMT, resting motor threshold; SICI, short-interval intracortical inhibition; TMS, transcranial magnetic stimulation; TS, test stimulus.
Introduction
Corticospinal excitability is modulated by a variety of sensory inputs, including auditory (Furubayashi et al., 2000), somatosensory (Tokimura et al., 2000), visual (Cantello
et al., 2000), and even gustatory (Mistry et al., 2006). In particular, somatosensory input has
often been given special prominence, in view of its direct and short latency inputs. A large proportion of motor cortex neurones recorded in non-human primates respond at short latency to somatosensory inputs (Cheney & Fetz, 1984), and such responses are likely to be involved in long-latency transcortical stretch and cutaneous reflexes in humans (Macefield et al., 1996).
In contrast, visual inputs are usually viewed as relatively indirect and weak, with only about 3% neurones in primate motor cortex responding to visual stimulation (Lamarre et al., 1983). In humans there have been relatively few direct investigations of the effects of visual input on motor cortex, but those that have been done suggest that relatively strong effects can be observed at surprisingly short latency. The earliest studies were made in patients with photic reflex myoclonus in whom flashes of light can evoke a generalised myoclonic jerk (Shibasaki & Neshige, 1987). In a series of investigations on 6 patients, Artieda and Obeso (Artieda & Obeso, 1993) concluded that the flash evoked an early wave of EEG activity in visual cortex starting 34 ms after the stimulus, and that this was followed by bilateral activation of motor cortex at 42 ms. This preceded the onset of muscle jerks in biceps by about 11 ms, consistent with rapid conduction down the corticospinal tract. The authors suggested that visual input was reaching the motor cortex from primary visual areas since transcranial magnetic stimulation (TMS) over the occiput during 1 Hz flash stimuli (to increase visuo-motor excitability) provoked a muscle twitch some 7 ms later than direct TMS over M1. A later study by Cantello et al (2000) in healthy volunteers followed up on these observations by using single pulses of TMS to assess the excitability of the motor cortex after a light flash. They found that excitability was reduced some 55-70 ms after the flash followed by facilitation at around 120 ms. Since there was no effect of the stimulus on measures of spinal cord excitability (H-reflex and F-waves) at a time corresponding to the early phase of inhibition, they suggested that it was due to modulation of motor cortex excitability by visual inputs. They pointed out that the response to a flash reaches visual cortex at about 40 ms, so that if a cortico-cortical pathway were involved from visual (V1) to motor cortex (M1), the transit time would be of the order of 15 ms.
Rapid access of visual input to motor areas of cortex is also evident from many reaction time studies. For example, O’Shea et al (2007) found that a visual stimulus could increase motor cortex excitability during a simple reaction time task as early as 50 ms after onset, and could influence connectivity between premotor and motor cortex after only 75 ms. Similarly Makin et al (2009) found that MEPs in hand muscles were reduced during a simple visuo-manual task if participants saw adistractor ball rapidly approaching a location near the hand. The minimal latency from onset of the movement was remarkably short at around 70 ms.
The present experiments were designed to investigate the possibility that inputs from primary visual areas may be one source of rapid input to M1 by using a “twin coil” TMS approach to test whether a conditioning pulse over the occiput influences the amplitude of the muscle twitches evoked from a later TMS pulse applied over M1. We tested connectivity at rest as well as during the warning period prior to a simple visual reaction time task in order to examine whether connectivity would be task-dependent.
Materials and Methods
Subjects.
A total of 16 healthy volunteers (7 women, 21–51 years old) participated in this study, although not all of them participated in all experiments. All subjects were right-handed based on the Edinburgh Handedness Inventory and gave written informed consent. Experiments were approved by the local Ethics Committee and were performed in accordance with the Declaration of Helsinki.
TMS
For paired-TMS we used two high-power Magstim 200 machines (Magstim,
Whitland, UK). The magnetic stimulus had a nearly monophasic pulse configuration with a rise time of ~ 100 µs, decaying back to zero over ~ 0.8 µs. The stimulators were connected to a figure-of-eight coil (outer winding diameter 70 mm).
Test stimuli
MEPs were recorded from the first dorsal interosseous (FDI) muscles using 9 mm diameter Ag-AgCl surface-cup electrodes, in a typical belly-tendon montage. Responses were
amplified by a Digitimer D360 device (Digitimer, Welwyn Garden City, UK). Filters were 20 Hz - 3 kHz, and the sampling rate was 10 kHz. The signal was then recorded by a PC using Signal software ver. 4.08 (Cambridge Electronic Devices, Cambridge, UK). The test coil was placed tangentially to the scalp at a 45° angle to the midline, to induce a posterior-anterior (PA) current flow across the central sulcus. The hand motor area of the left M1 was defined as the point where stimulation consistently evoked the largest MEP. We defined the resting motor threshold (RMT) as the lowest intensity that evoked 5 small responses (~50 µV) in the relaxed FDI muscle in a series of 10 stimuli (Rossini et al., 1994). The intensity of the TS was finally adjusted to evoke a MEP of ~ 1 mV peak-to-peak amplitude in the relaxed right FDI.
Experiment 1 (Pilot study, n = 10).
Eyes were open. The TS was delivered as described above. The centre of the
conditioning coil was placed over Oz (10–20 EEG system) with its handle pointing upward, on the inion–nasion line, tangential to the skull. The intensity of the CS was arbitrarily set at 90% RMT. Interstimulus intervals (ISIs) were 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 35 and 40 ms. Thirteen conditions were randomly intermingled: TS alone and CS plus TS at each ISI. Fifteen responses were collected for TS and 12 responses for CS plus TS, for a total of 159 trials. There was a 5 s (±20%) intertrial interval. For each trial we measured the average peak-to-peak MEP amplitude. The conditioned MEP was expressed as a percentage of the unconditioned MEP size.
Experiment 2 (n =11).
Paired-TMS stimulation was conducted as in Exp. 1, but the centre of the conditioning coil was placed over the phosphene hot spot. To assess the latter, and the phosphene threshold (PT) according to the method of Stewart et al. (Stewart et al., 2001), subjects wore a blindfold and a cap whilst seated in a comfortable chair in a dim light room. Three points were marked over the occipital midline 2, 3 and 4 cm above the inion. The coil handle pointed upwards and was parallel to the subject’s spine. The coil centre was first positioned 2 cm above the inion, then moved anteriorly across the marks, to determine the best site to elicit phosphenes (“hot spot”). Shocks were initially applied at 60% of the stimulator output and at a maximum frequency of 0.2 Hz. The subject was asked about the presence of phosphenes immediately after each pulse. If a phosphene was reported 5 or more times out of 10, the pulse intensity was reduced by steps of 5%, then shocks were repeated
another 10 times. This protocol progressed till no phosphene was reported. The minimum intensity at which the subject perceived a phosphene 5 times out of 10 was the PT. If the initial intensity of 60% was ineffective, it was increased by steps of 5% maximum power, till phosphenes appeared. One subject was excluded since he reported no phosphenes. The intensity of the CS was adjusted to be 80% PT or 90% PT. ISIs were 12, 15, 18, 21, 24, 27, 30, 35 and 40 ms. ISIs between 3 and 9 ms were not studied because they were ineffective in Exp. 1. There were two sessions: one with eyes open and another with eyes closed.
Experiment 3 (n =8).
We then studied the effects of changing the CS site, in a setting otherwise identical to Exp. 2. Conditioning stimuli, whose intensity was 90% PT, were comparatively applied to the phosphene hot spot or to a site 3 cm lateral to Pz on the right side. The subject eyes were open.
Experiment 4 (n=6).
The protocol described in Exp. 2 was then repeated during a visuo-motor RT task. We hypothesized that a physiologically relevant connectivity would show time-specific changes in such a context. We used a task similar to that of Touge et al (1998). Subjects sat relaxed in a chair with their right forearm lying comfortably on a pillow and their right hand on a button box. Eyes were open. Surface EMG was recorded from the FDI, APB (abductor pollicis brevis) and ADM (abductor digiti minimi) muscles. We ensured that there was no EMG activity at baseline. A black screen was placed in front of the subjects at a distance of 50 cm, which carried two light-emitting diodes (LEDs) separated by 1.5 cm. The red LED was the warning signal (WS) and the green LED was the response signal (RS). Subjects were instructed to use the WS to prepare for the upcoming response, then to contract their right FDI muscle as quickly as possible, to press the button with their right index finger, as soon as they saw the RS. Each trial began with a WS followed by a RS, given randomly 600±50 ms later. The intertrial interval was 5 s (±20%). We had two randomized sessions separated by at least one week. In each session we measured the effects of the CS on TS while subjects were at rest, outside of the reaction time task. CS was 90% PT. ISIs of 18 and 40 ms (the most effective in previous experiments) were randomly intermingled. Subjects also performed 4 blocks of the RT task. Each block had 4 conditions that were randomised within the block. Condition 1: subjects received a WS, followed 600±50 ms later by a RS, to which they had to react as fast as possible. Condition 2: a TS alone given at -300, -150, -50 or +50 ms
(depending on the block, see below). Condition 3: same as condition 2, but the TS was preceded by a CS with an ISI of either 18 or 40 ms (depending on the block, see below). Condition 4: a TS alone was given in the intertrial interval (Figure 1). Thirty trials were recorded for each condition for a total of 120 trials. In one of the experimental sessions, the four trial blocks were: (1) TS at -300 ms, CS 18 ms before test; (2) TS at -150 ms, CS at 18 ms; (3) TS at -300 ms, CS at 40 ms before test; (4) TS at -150 ms, CS at 40 ms. The other experimental session contained TS at -50 ms and +50 ms. Before each session, at least 50 practice trials were given.
Experiment 5 (n=6)
We re-tested the paired-TMS protocol during an auditory RT task. The subjects, general settings and conditions were the same as in Exp. 4. We used however an auditory RT task, where the first tone (500 Hz, 50 ms) was the WS, while the second tone (1000 Hz, 50 ms) was the RS. Also, we restricted our timings to TS at -50 ms (i.e. just prior to the RS) and ISI between CS and TS to 40 ms since these parameters had produced large effects in
experiment 4.
Experiment 6 (n=6)
In six subjects, we investigated the effects of a CS over the visual cortex on short interval intracortical inhibition (SICI) in the left M1 (Kujirai et al., 1993). We used three high-power Magstim 200 machines. The first conditioning stimulus (CS1) was delivered with an intensity of 90% PT over the phosphene hot spot, and the second one (CS2) over the left M1. The intensity of CS2 was set to the relatively low value of 70% active motor threshold (AMT), to avoid floor effects on the percentage SICI. AMT was defined as the lowest intensity that evoked five small responses (about 100 µV) in a series of ten stimuli when the subject made a 10% of the maximum voluntary contraction of the right FDI. The ISIs between CS1 and CS2 were 18 and 40 ms, whilst the ISI between CS2 and TS was 2.2 ms. First we tested the effects on the test MEP of giving CS1 alone (with an ISI of 40 and 18 ms) or CS2 alone (baseline SICI). Then, the intensity of the TS was re-adjusted, so that when CS1+TS was applied, the combined effect would elicit a MEP of ~ 1 mV. Finally, CS1+CS2 were delivered at ISIs of 18 and 40 ms. Fifteen trials were recorded for each condition.
All data were expressed as mean ± standard error of the mean (SEM). Student’s paired t tests (two-tailed) was used to compare mean RMT with eyes open and closed obtained from all the participants. Spearman's rho was applied to study the correlation between motor and phosphene threshold. In general, the effects of the CS on the MEP
amplitude were analysed with separate one-way ANOVAs for any given stimulation intensity and eyes state, with “ISI” (TS alone, CS plus TS at various ISIs) as the main factor. A
significant main effect in these ANOVAs was followed by post hoc tests with Dunnett corrections. Based on the conditions of the various experiments, we performed preliminary two or three-ways repeated-measures (rm) ANOVAs that accounted for the various factors to be analysed. Supplementary ANOVAs or rmANOVAs were finally carried out as dictated by the specific experiment, to assess the effects of additional confounders, e.g. in Exp. 4, a two-way rmANOVA explored the “time” (Figure 1) x “ISI” interactions. Mauchley’s test
examined for sphericity. The Greenhouse-Geisser correction was used for nonspherical data. Occasionally, two-tailed paired Student t tests were used (Exp. 6). A p value < 0.05 was considered significant. Data were analysed using software (SPSS v. 19.0 for Windows; SPSS Inc.).
Results
Mean RMT with eyes open was 41.1% (range, 30–52%), the same as with eyes closed (40.6%; range 30-53%) (Student t = 1.472, p = 0.175). Mean PT was 62.5% (range, 42–76%). Motor and phosphene thresholds did not correlate (Spearman's rho = -0.175, p = 0.566 with eyes open; rho = -0.161 p = 0.617 with eyes closed).
Experiment 1 (Conditioning MEPs with stimuli over Oz at rest).
Figure 2 shows the mean effect of conditioning V1 on MEPs evoked from M1 at a range of ISIs. A one-way ANOVA revealed a significant main effect of “ISI” (F (12, 105) = 2.546, p = 0.005). Post hoc analysis disclosed that the MEPs were significantly reduced at ISI 18 ms (p = 0.049) with a borderline reduction at ISI 24 ms as well (p = 0.085).
Experiment 2 (Conditioning MEPs with stimuli over the phosphene hot spot at rest).
In this experiment, the CS was placed over the phosphene hot spot. The effect of two different intensities of CS was measured on MEPs evoked from the ipsilateral M1 with eyes open or closed throughout the testing (Figure 3). A preliminary three-way rmANOVA
showed a significant main effect of “ISI” (F (8, 320) = 9.653, p < 0.001), but no effect of “eye state” (F (1, 40) = 0.531, p = 0.470) or “intensity” (F (1, 40) = 0.355, p = 0.555) and no significant interactions (p > 0.05). Thus the time course of MEP suppression was the same at each intensity of CS and was unaffected by eye closure. The graphs also indicate the ISIs in each state where post hoc testing revealed significant (p < 0.05) effects compared with control (Figure 3). Because ISIs of 18 and 40ms were effective in all states these two intervals were then used in experiments 4-6.
Experiment 3 (Changing the site of the conditioning stimulus).
To confirm that the effect of the CS was spatially specific we compared the effect of conditioning over the phosphene hot spot with conditioning over a point 3cm lateral to Pz. Figure 4 shows that stimulation over the parietal site at this intensity had no effect whereas there was clear MEP suppression if the CS was over V1. A two-way rmANOVA showed a significant main effect of “stimulation site” (F (1, 7) = 37.517, p < 0.001), as well as a significant interaction between “stimulation site” and “ISI” (F (8, 56) = 2.475, p = 0.023), indicating that the time course of the effect on MEPs differed between sites. Follow up one-way ANOVAs revealed a significant main effect of ISI (F (9, 63) = 4.734, p < 0.001) at the phosphene hot spot but no effect of ISI over the parietal site (F (9, 63) = 1.647, p = 0.121). On post hoc analysis, the size of the MEP conditioned from V1 was significantly reduced at ISI 18 ms (p = 0.001), 21 ms (p = 0.014) and 40 ms (p = 0.002). No subject reported
phosphenes after the control (parietal) stimulus.
Experiment 4 (Visuomotor functional connectivity during a visual RT task)
We next tested whether the effect of the CS varied during the course of a warned simple visual reaction time task. MEPs were conditioned by stimulation over the phosphene hot spot during the warning interval prior to the onset of the RS, and at 50 ms following the RS prior to onset of movement. The effects were compared with those seen at complete rest outside the reaction task. MEPs to the M1 stimulus given alone were the same at rest as at all intervals tested during the task (one-way rmANOVA, first session of task: (F (3, 15) = 0.656,
p = 0.591); second session (F (3, 15) = 1.617, p = 0.227) (Figure 5).
Figure 6 plots the size of the conditioned MEP as a percent of the test MEP alone for the two ISIs between CS and TS (18 and 40 ms). There are five bars for each ISI
corresponding to suppression at rest and at -300, -150, -50 and +50 (with respect to the time of the RS) during the reaction task. The percent suppression of MEP at an ISI of 18 ms was
unchanged during the task whereas suppression at ISI = 40 ms gradually shifted to facilitation around the time of the RS.
This was confirmed by a two-way rmANOVA showing a significant main effect of “time” (F (4, 20) = 18.290, p < 0.001), “ISI” (F (1, 5) = 20.066, p = 0.007) and a significant interaction “time” x “ISI” (F (4, 20) = 7.234, p = 0.001). Follow up one-way ANOVAs showed no effect of “time” with an ISI = 18 ms (F (4, 20) = 0.607, p = 0.662), whereas there was a significant effect at ISI = 40 ms (F (4, 20) = 21.355, p < 0.001). Post hoc analysis showed that the conditioned MEP was significantly larger 150 ms before (p = 0.004), and both 50 ms before and after the RS (p < 0.001).
Experiment 5 (Visuomotor functional connectivity during an auditory RT task)
In the visual task the CS (ISI = 40 ms) facilitated the conditioned MEP 50 ms prior to the RS. In the same subjects we compared this with the effect at the same timing in an
auditory reaction task. The unconditioned MEP at rest was the same as during the visual and auditory task (50 ms before the RS) (F (3,15) = 0.848, p = 0.489). Figure 7 shows that the CS suppressed the MEP by a similar amount when subjects were tested at rest in the auditory and visual sessions. However, during performance of the auditory task (-50 ms) there was no longer any effect of the CS on the TS whereas in the visual task it was facilitated. A one-way rmANOVA on the data confirmed that the effect of the CS differed between the four
conditions (F (3, 15) = 23.647, p < 0.001). Follow-up analysis showed that although there was a significant difference between the effect at rest and at the -50 ms time points in both tasks (visual, p < 0.001; auditory, p = 0.003), the effect was larger in the visual task compared with the auditory task (p = 0.029).
Experiment 6 (Effects on SICI)
A CS over the phosphene hot spot increased the amount of the baseline SICI (from 76.6% to 57.7%) (Student t = 4.643, p = 0.006) at an ISI of 40 ms, but not at 18 ms (t = 0.119, p = 0.910) (Figure 8). As a result of intensity re-adjustment, the test MEP size was 0.93 ± 0.1 mV (ISI 40 ms) and 1.07 ± 0.1 (ISI 18 ms), i.e. it was not statistically different from the unconditioned MEP (1.16 ± 0.1 mV) (F (2, 10) = 1.324, p = 0.309).
The present data show that TMS over the occipital region affects excitability of M1 when tested 18-40 ms later. The effect was not caused by the auditory click made by the coil when discharged (Furubayashi et al., 2000), as it was no longer present when the site of stimulation was moved 3 cm lateral to Pz. Since the TMS coil was located over the optimal point to elicit stationary phosphenes (Afra et al., 1998; Stewart et al., 2001; Franca et al., 2006) (Figure 9), and used an intensity below phosphene threshold, we suggest that the effect depends on activation of primary visual cortex (V1).The effect was present at both 80% and 90% phosphene threshold (PT) but was not significantly influenced by whether the eyes were open or closed. At first sight this seems unexpected since transient removal of vision
increases the excitability of V1, as reflected by larger amplitude early components of the flash evoked EEG potential (Cantello et al., 2011). However, it may be that although the visual input to V1 is facilitated there is little effect of vision on the output from V1 that is the target of the TMS conditioning pulse in the present experiment. It has also been noted that blindfolding increases excitability of M1, as tested by its effect on the amplitude of TMS-evoked muscle twitches (Leon-Sarmiento et al., 2005). However, this again may have little relevance to the effectiveness of inhibitory inputs evoked by TMS over V1 as we did not detect any change of RMT or baseline MEP between open and closed eyes. We are less certain why the responses to conditioning stimuli of 80% and 90% PT were similar. It seems possible that this was due to a lack of statistical power, given the tendency for more
inhibition to occur at 90% PT whether the eyes were open or closed.
Our results confirm the evidence reviewed in the Introduction that activity in visual cortex can modulate corticospinal excitability at short latency in subjects at rest. One of the limits of previous approaches is that they used natural visual stimuli and there is some
uncertainty about the precise time at which these arrive in visual cortex. Most studies indicate the first occipital visual evoked potentials begin around 35-40 ms (ffytche et al., 1995), while intracranial electrodes recorded a latency of about 31-33 ms (Ducati et al., 1988). Using these figures, the earliest TMS effect at ISI = 18 ms is compatible with the data on flash evoked suppression of MEPs noted by Cantello and colleagues (Cantello et al., 2000; O'Shea et al., 2007; Makin et al., 2009) but slightly later than the very rapid (7 ms) visuo-motor
connectivity described in photic reflex myoclonus (Nakashima et al., 1985; Shibasaki & Neshige, 1987; Artieda & Obeso, 1993; Kanouchi et al., 1997). The shorter occipitomotor conduction time in the patients might well be explained by a pathological exaggeration of the normal physiological mechanism, resulting in a shift from inhibition to excitation of the motor cortex. A similar mechanism might explain the spreading of the epileptic discharge
from the hyperexcitable visual cortex to the motor cortex in photosensitive idiopathic epilepsies (Strigaro et al., 2012).
The later phase of interaction at ISI = 40 ms is compatible with the earliest signs of visual effect on motor cortex excitability described in a number of behavioural studies (e.g. Makin et al, 2009). In addition, Suppa et al (2013) showed that it was possible to induce long-term potentiation (LTP) and depression (LTD)-like plasticity in the primary motor cortex in healthy humans after repetitive pairing of a patterned visual stimulus and a TMS stimulus at specific time intervals around the latency of the P100 evoked potential. These varied between 40 and 140 ms after the individual P100 latency (i.e. between 140 and 240 ms after onset of the visual stimulus) (Suppa et al., 2013) and are therefore longer than the ISIs we deal with in the present paper. However, the data are important because they suggest that there may be a range of later pathways available that may have important behavioural effects.
Studies on primates confirm the existence of relatively rapid connections between visual and motor areas of cortex. Indeed, visual input can activate wide areas of cerebral cortex not directly involved in vision (i.e. the premotor cortex, supplementary motor area, prefrontal cortex, frontal ocular fields (Fadiga et al., 2000). Ledberg et al (2007) analysed the event-related local field potentials (LFP) recorded simultaneously from multiple visual, motor, and frontal areas in monkeys performing a visuomotor discrimination task. They confirmed (Boussaoud & Wise, 1993; Thut et al., 2000; Saron et al., 2001; Foxe & Simpson, 2002) the occurrence of early activation in motor, premotor, and prefrontal areas following stimulus presentation, which in some cases was as early as in striate cortex (Ledberg et al., 2007).
In the last decade, diffusion tensor imaging (DTI) techniques (Catani et al., 2002) and anatomical dissection studies (Martino et al., 2010; Sarubbo et al., 2011) demonstrated the existence in humans of the inferior fronto-occipital fascicle (IFOF), a long associative bundle connecting the occipital cortex and other posterior areas to the frontal lobe (Martino et al., 2010). In particular, the IFOF might constitute a direct efferent pathway from the occipital associative cortex, by rapidly transmitting an initially processed visual information to the prefrontal regions (Martino et al., 2010). Its functional role is not yet clear, but it might take part in a ‘‘multimodal integration’’ among these distant brain regions (Sarubbo et al., 2011). It might thus account for the fast conduction of V1 stimuli to the motor areas as revealed by our experiment.
Most long range cortico-cortico connections are thought to be excitatory, as in the transcallosal pathway. The fact that we obtain an overall inhibitory effect in the present
experiments would therefore be compatible with the idea that these excitatory projections synapse onto inhibitory interneurones in M1 that suppress corticospinal excitability. This is supported by our findings that a CS over the visual cortex increased SICI in the left M1, at least for ISI = 40 ms (not 18 ms). SICI describes the local interaction between two stimuli applied to M1 through the same coil. A small CS suppresses the response to a TS given 1-5 ms later. Pharmacological interventions show that this local interaction is due to activity in GABAa-ergic inhibitory interneurones. Thus the fact that SICI is made more effective by stimulation over visual cortex suggest that occipital input has access to inhibitory circuits in M1 and that this may contribute to the MEP suppression we have described. Visuo-motor suppression at 18 ms presumably does not depend on activity in the same set of interneurones since it has no effect on SICI. However there are a number of possibilities that can be tested with TMS methods, including a GABAb-ergic system (tested with the long interval
intracortical inhibition paradigm) and a further pathway modulated by cholinergic input (tested with short afferent inhibition). Further work could tease apart these possibilities. At the present time we conclude that the two phases of inhibition are caused by activity in two distinct pathways.
Visual information is essential to the preparation, execution and on-line control of voluntary movement. To assess its potential physiological role, we examined visuomotor connectivity during a visual RT task using ISIs of 18 and 40 ms since they produce the most consistent inhibitory effects. The task had no effect on MEP suppression at ISI = 18 ms at any of the time points studied during the task. This was not true for ISI = 40 ms. The
inhibitory effect at rest (MEP reduced by 30-40%) gradually reversed into facilitation during movement preparation. Facilitation appeared to begin about 150 ms prior to the RS and was very clear at +50 ms (MEP increased by 40-50%). This contrasts with the results in an equivalent auditory reaction task. The usual visuo-motor suppression observed at rest was absent 50 ms prior to the RS, but there was no frank facilitation of the MEP as in the visual task. We suggest that rapid visuomotor connectivity is suppressed during an auditory task but becomes facilitatory during a visual task, perhaps improving access of visual input to motor areas. It is unclear why connectivity at ISI = 18 ms was unaffected in the visual reaction task. Nevertheless, the finding does confirm the conclusion that these two effects are mediated by quite separate pathways.
During the RT tasks we saw no significant changes in the unconditioned MEP at the time intervals we studied. In some previous studies, the MEP has been suppressed in the interval between the WS and RS (Hasbroucq et al., 1997; Touge et al., 1998; Davranche et
al., 2007). However, suppression is best observed when the WS-RS interval is constant and
subjects can anticipate precisely when the RS is about to be delivered (Touge et al., 1998). In the present task the timing of the RS was not predictable since it was randomised to come 550-650 ms after the WS.
Conclusions
Our findings support the existence of physiologically relevant occipitomotor connections, which can be activated by means of TMS. They may contribute to rapid
integration of visual input into motor tasks as well as being involved in certain types of visual epilepsy.
Author contributions
G. S., J.C.R., D.R. and R.C. conceived and designed the experiments. G.S., J.C.C., L.M. and M.D. collected the data. All authors participated in the analysis and interpretation of the data and in the drafting of the manuscript. All authors approved the final version of the
manuscript.
Acknowledgements
G.S. is funded by the Fondazione Veronesi Young Investigator Programme 2013. Conflict of interest
Legend of Figures. Figure 1.
The setting of Experiment 4. In condition 1, subjects received a warning signal (WS), followed 600±50 ms later by a response signal (RS), after which they had to react as fast as possible. In condition 2, a test stimulus (TS) alone given at one of four different “times” (-300, -150, -50 or +50 ms). In condition 3, same as condition 2, but the TS was preceded by a conditioning stimulus (CS) with an ISI of either 18 or 40 ms. In condition 4 a TS alone was given in the intertrial interval.
Figure 2.
Effects of a conditioning stimulus (CS) applied over Oz at 90% relaxed motor threshold (RMT) on the MEPs obtained by a test stimulus (TS) applied over the left M1, while subjects were at rest with eyes open. CS preceded TS by different ISIs, ranging from 3 to 40 ms. Amplitude of MEPs (mV) is normalized and expressed as a percentage of control test. Errors bars indicate SEM. Asterisks indicate a p value < 0.05 on post hoc analysis.
Figure 3.
Effects of a conditioning stimulus (CS) applied over the phosphene hot spot at different intensities (80% or 90% PT) and eye states (eyes open or closed) on the test MEPs with subjects at rest. Amplitude of MEPs (mV) is normalized and expressed as a percentage of control. Errors bars indicate SEM. Asterisks indicate a p value < 0.05 on post hoc analysis. Figure 4.
Effects of changing the location of the conditioning stimulus (CS, 90% of the phosphene threshold, PT) on the test MEPs with subjects at rest. Grey line: CS applied to a scalp site 3 cm lateral to Pz on the right side. Black line: CS applied to the phosphene hot spot.
Amplitude of MEPs (mV) is normalized and expressed as percentages of control. Errors bars indicate SEM. Asterisks indicate a p value < 0.05 on post hoc analysis.
Figure 5.
Unconditioned MEPs at rest, in the intertrial interval and at all intervals tested during the simple visual reaction time task. Left, first session. Right, second session. Errors bars indicate SEM.
Figure 6.
Effects of the conditioning stimulus (CS, phosphene hotspot) on the test MEP amplitude at rest and at different times during the behavioural task (-300, -150, - 50 and +50 ms). Left, ISI 18 ms. Right, ISI 40 ms. Amplitude of MEPs (mV) is normalized and expressed as
percentage of control. Errors bars indicate SEM. Figure 7.
Effects of the conditioning stimulus (CS, phosphene hotspot) with an ISI of 40 ms on the MEP amplitude at rest and during a visual and an auditory reaction task 50 ms before the response signal (RS). Amplitude of MEPs (mV) is normalized and expressed as a percentage of control. Errors bars indicate SEM.
Figure 8.
Comparison of the effects on short-interval intracortical inhibition (SICI) of conditioning stimuli applied over the visual cortex with an ISI of 18 and 40 ms. Errors bars indicate SEM. Figure 9.
MRI reconstruction of a single subject. The red mark indicates the orientation of the magnetic field at the phosphene hot spot (striate cortex). The anterior green dot is at the hand area of the left motor cortex.
References
Afra J, Mascia A, Gerard P, Maertens de Noordhout A & Schoenen J. (1998). Interictal cortical excitability in migraine: a study using transcranial magnetic stimulation of motor and visual cortices. Ann Neurol 44, 209-215.
Artieda J & Obeso JA. (1993). The pathophysiology and pharmacology of photic cortical reflex myoclonus. Ann Neurol 34, 175-184.
Boussaoud D & Wise SP. (1993). Primate frontal cortex: neuronal activity following attentional versus intentional cues. Exp Brain Res 95, 15-27.
Cantello R, Civardi C, Cavalli A, Varrasi C & Vicentini R. (2000). Effects of a photic input on the human cortico-motoneuron connection. Clin Neurophysiol 111, 1981-1989.
Cantello R, Strigaro G, Prandi P, Varrasi C, Mula M & Monaco F. (2011). Paired-pulse flash-visual evoked potentials: new methods revive an old test. Clin Neurophysiol 122, 1622-1628. Catani M, Howard RJ, Pajevic S & Jones DK. (2002). Virtual in vivo interactive dissection of white
matter fasciculi in the human brain. Neuroimage 17, 77-94.
Cheney PD & Fetz EE. (1984). Corticomotoneuronal cells contribute to long-latency stretch reflexes in the rhesus monkey. J Physiol 349, 249-272.
Davranche K, Tandonnet C, Burle B, Meynier C, Vidal F & Hasbroucq T. (2007). The dual nature of time preparation: neural activation and suppression revealed by transcranial magnetic stimulation of the motor cortex. Eur J Neurosci 25, 3766-3774.
Ducati A, Fava E & Motti ED. (1988). Neuronal generators of the visual evoked potentials:
intracerebral recording in awake humans. Electroencephalogr Clin Neurophysiol 71, 89-99. Fadiga L, Fogassi L, Gallese V & Rizzolatti G. (2000). Visuomotor neurons: ambiguity of the discharge
or 'motor' perception? Int J Psychophysiol 35, 165-177.
ffytche DH, Guy CN & Zeki S. (1995). The parallel visual motion inputs into areas V1 and V5 of human cerebral cortex. Brain 118 ( Pt 6), 1375-1394.
Foxe JJ & Simpson GV. (2002). Flow of activation from V1 to frontal cortex in humans. A framework for defining "early" visual processing. Exp Brain Res 142, 139-150.
Franca M, Koch G, Mochizuki H, Huang YZ & Rothwell JC. (2006). Effects of theta burst stimulation protocols on phosphene threshold. Clin Neurophysiol 117, 1808-1813.
Furubayashi T, Ugawa Y, Terao Y, Hanajima R, Sakai K, Machii K, Mochizuki H, Shiio Y, Uesugi H, Enomoto H & Kanazawa I. (2000). The human hand motor area is transiently suppressed by an unexpected auditory stimulus. Clin Neurophysiol 111, 178-183.
Hasbroucq T, Kaneko H, Akamatsu M & Possamai CA. (1997). Preparatory inhibition of cortico-spinal excitability: a transcranial magnetic stimulation study in man. Brain Res Cogn Brain Res 5, 185-192.
Kanouchi T, Yokota T, Kamata T, Ishii K & Senda M. (1997). Central pathway of photic reflex myoclonus. J Neurol Neurosurg Psychiatry 62, 414-417.
Kujirai T, Caramia MD, Rothwell JC, Day BL, Thompson PD, Ferbert A, Wroe S, Asselman P & Marsden CD. (1993). Corticocortical inhibition in human motor cortex. J Physiol 471, 501-519.
Lamarre Y, Busby L & Spidalieri G. (1983). Fast ballistic arm movements triggered by visual, auditory, and somesthetic stimuli in the monkey. I. Activity of precentral cortical neurons. J
Neurophysiol 50, 1343-1358.
Ledberg A, Bressler SL, Ding M, Coppola R & Nakamura R. (2007). Large-scale visuomotor integration in the cerebral cortex. Cereb Cortex 17, 44-62.
Leon-Sarmiento FE, Bara-Jimenez W & Wassermann EM. (2005). Visual deprivation effects on human motor cortex excitability. Neurosci Let 389, 17-20.
Macefield VG, Rothwell JC & Day BL. (1996). The contribution of transcortical pathways to long-latency stretch and tactile reflexes in human hand muscles. Exp Brain Res 108, 147-154. Makin TR, Holmes NP, Brozzoli C, Rossetti Y & Farne A. (2009). Coding of visual space during motor
preparation: Approaching objects rapidly modulate corticospinal excitability in hand-centered coordinates. J Neurosci 29, 11841-11851.
Martino J, Brogna C, Robles SG, Vergani F & Duffau H. (2010). Anatomic dissection of the inferior fronto-occipital fasciculus revisited in the lights of brain stimulation data. Cortex 46, 691-699. Mistry S, Rothwell JC, Thompson DG & Hamdy S. (2006). Modulation of human cortical swallowing
motor pathways after pleasant and aversive taste stimuli. Am J Physiol Gastrointest Liver
Physiol 291, G666-671.
Nakashima K, Araga S & Takahashi K. (1985). Electrophysiological studies of myoclonic jerks provoked by photic stimulation. Acta Neurol Scand 71, 401-407.
O'Shea J, Sebastian C, Boorman ED, Johansen-Berg H & Rushworth MF. (2007). Functional specificity of human premotor-motor cortical interactions during action selection. Eur J Neurosci 26, 2085-2095.
Rossini PM, Barker AT, Berardelli A, Caramia MD, Caruso G, Cracco RQ, Dimitrijevic MR, Hallett M, Katayama Y, Lucking CH & et al. (1994). Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical
application. Report of an IFCN committee. Electroencephalogr Clin Neurophysiol 91, 79-92. Saron CD, Schroeder CE, Foxe JJ & Vaughan HG, Jr. (2001). Visual activation of frontal cortex:
segregation from occipital activity. Brain Res Cogn Brain Res 12, 75-88.
Sarubbo S, De Benedictis A, Maldonado IL, Basso G & Duffau H. (2011). Frontal terminations for the inferior fronto-occipital fascicle: anatomical dissection, DTI study and functional
considerations on a multi-component bundle. Brain Struct Funct.
Shibasaki H & Neshige R. (1987). Photic cortical reflex myoclonus. Ann Neurol 22, 252-257. Stewart LM, Walsh V & Rothwell JC. (2001). Motor and phosphene thresholds: a transcranial
magnetic stimulation correlation study. Neuropsychologia 39, 415-419.
Strigaro G, Prandi P, Varrasi C, Monaco F & Cantello R. (2012). Defective visual inhibition in photosensitive idiopathic generalized epilepsy. Epilepsia.
Suppa A, Li Voti P, Rocchi L, Papazachariadis O & Berardelli A. (2013). Early Visuomotor Integration Processes Induce LTP/LTD-Like Plasticity in the Human Motor Cortex. Cereb Cortex.
Thut G, Hauert CA, Blanke O, Morand S, Seeck M, Gonzalez SL, Grave de Peralta R, Spinelli L, Khateb A, Landis T & Michel CM. (2000). Visually induced activity in human frontal motor areas during simple visuomotor performance. Neuroreport 11, 2843-2848.
Tokimura H, Di Lazzaro V, Tokimura Y, Oliviero A, Profice P, Insola A, Mazzone P, Tonali P & Rothwell JC. (2000). Short latency inhibition of human hand motor cortex by somatosensory input from the hand. J Physiol 523 Pt 2, 503-513.
Touge T, Taylor JL & Rothwell JC. (1998). Reduced excitability of the cortico-spinal system during the warning period of a reaction time task. Electroencephalogr Clin Neurophysiol 109, 489-495.
