Brain Stimulation xxx (2017) 1e10

Contents lists available at ScienceDirect

Brain Stimulation journal homepage: http://www.journals.elsevier.com/brain-stimulation

TMS over posterior parietal cortex disrupts trans-saccadic visual stability
re se Collins a, *, Pierre O. Jacquet a, b, c The Laboratoire Psychologie de la Perception, Universit
e Paris Descartes & CNRS, 45 rue des Saints-P eres 75006 Paris, France Laboratoire de Neurosciences Cognitives (LNC), D
epartement d’Etudes Cognitives, INSERM U960, Ecole Normale Sup
erieure, PSL Research University, 75005 Paris, France c Institut Jean Nicod, D
epartement d’Etudes Cognitives, CNRS UMR8129, Ecole Normale Sup
erieure, PSL Research University, 75005 Paris, France a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 July 2017 Received in revised form 24 November 2017 Accepted 26 November 2017 Available online xxx

Background: Saccadic eye movements change the retinal location of visual objects, but we do not experience the visual world as constantly moving, we perceive it as seamless and stable. This visual stability may be achieved by an internal or efference copy of each saccade that, combined with the retinal information, allows the visual system to cancel out or ignore the self-caused retinal motion. Objective: The current study investigated the underlying brain mechanisms responsible for visual stability in humans with online transcranial magnetic stimulation (TMS). Methods: We used two classic tasks that measure efference copy: the double-step task and the in-flight displacement task. The double-step task requires subjects to make two memory-guided saccades, the second of which depends on an accurate internal copy of the first. The in-flight displacement task requires subjects to report the relative location of a (possibly displaced) target across a saccade. In separate experimental sessions, subjects participated in each task while we delivered online 3-pulse TMS over frontal eye fields (FEF), posterior parietal cortex, or vertex. TMS was contingent on saccade execution. Results: Second saccades were not disrupted in the double-step task, but surprisingly, TMS over FEF modified the metrics of the ongoing saccade. Spatiotopic performance in the in-flight displacement task was altered following TMS over parietal cortex, but not FEF or vertex. Conclusion: These results suggest that TMS disrupted eye-centered position coding in the parietal cortex. Trans-saccadic correspondence, and visual stability, may therefore causally depend on parietal maps. © 2017 Elsevier Inc. All rights reserved.

Keywords: Visual stability Saccade TMS

Introduction Our phenomenological experience of the visual world is stable and seamless. But the input to the visual system is constantly changing: as the eyes move around a visual scene, retinal coordinates vary. How visual stability is achieved has interested philosophers and psychologists for centuries, and has recently attracted renewed neurophysiological exploration [1]. One hypothesis about how stability arises is that an “extraretinal” (ER) signal cancels out retinal displacements due to selfmovement. Evidence for such a signal comes from the doublestep task [2] and the in-flight displacement task [3]. The doublestep task requires participants to make a series of two saccades to

* Corresponding author. E-mail address: [email protected] (T. Collins).

memorized targets; correctly executing the second saccade requires information about the first. In the in-flight displacement task, participants make a single saccade to a target that is displaced while the saccade is in mid-flight, and then report the direction of displacement. This task requires ER information in order to realign the pre- and post-saccadic views of the target. Neurons in a pathway from the superior colliculus to the frontal eye fields (FEF) via medio-dorsal thalamus carry a signal about saccade amplitude and direction that becomes available to the FEF before the actual eye movement takes place. Reversible lesions in the thalamic relay cause deficits in both double-step and in-flight displacement tasks suggestive of behavior expected in the absence of appropriate ER information [4e6]. The signal carried in the colliculo-frontal pathway is involved in realigning FEF neuronal receptive fields across eye movements [7], and the reafferent visual response of FEF neurons differs between targets that appear in the receptive field because of a saccade or because of target

https://doi.org/10.1016/j.brs.2017.11.019 1935-861X/© 2017 Elsevier Inc. All rights reserved.

Please cite this article in press as: Collins T, Jacquet PO, TMS over posterior parietal cortex disrupts trans-saccadic visual stability, Brain Stimulation (2017), https://doi.org/10.1016/j.brs.2017.11.019

2

T. Collins, P.O. Jacquet / Brain Stimulation xxx (2017) 1e10

displacement [8]. Patients with focal thalamic lesions show deficits that mirror those observed in reversible inactivation studies in monkeys [9,10]. In healthy subjects, offline theta-burst transcranial magnetic stimulation (rTMS) is thought to temporarily interfere with brain activity. Delivered over the FEF, it leads to deficits in the in-flight displacement task [11] suggestive of perturbed ER information. Post-parietal cortex (PPC) may also be important for ER signals. Neuropsychological studies reported deficits in saccade tasks compatible with the involvement of PPC in ER signals [12e15] but the often large extent of the cortical lesions and relatively small sample sizes make it difficult to pinpoint a more specific locus. Imaging studies also point towards involvement of the PPC in ER signals: PPC has an eye-centered organization that is updated when the eyes move [16,17] according to saccadic goals and not to the coordinates of the visual stimulus [18]. Furthermore, some PPC neurons show predictive remapping of their receptive fields: just prior to a saccade, RFs become responsive to stimuli in the future retinotopic location [12,19]. This means that the PPC receives information about saccades and updates retinotopic maps based on this signal. Taken together, these studies point to the importance of FEF and PPC for ER signals in visuo-motor tasks. To date, no study has used TMS to contrast the roles of FEF and PPC in humans in the doublestep and in-flight displacement tasks within the same subjects. The current study set out to do just that. In both tasks, one train of three TMS pulses was delivered right after the initiation of a saccade, allowing us to measure the behavioral consequences directly caused by the stimulation of specific neural populations in FEF and PPC. Materials and methods Subjects Seventeen subjects were recruited from Claude Bernard University of Lyon (France) and gave their written informed consent to participate in both the stimuli double-step task and the in-flight displacement task. They received payment for their participation. All participated in the double-step task (9 females; median age ¼ 29, range ¼ 20e42) and of these, 13 also participated in the in-flight displacement task (7 females; median age ¼ 29, range ¼ 20e42). All were right-handed, had normal vision and wore neither glasses nor contact lenses (for ease with eye movement recording). The experimental protocol was performed with approval of the regional ethics committee (CPP SUD EST IV) and in accordance with the Declaration of Helsinki [20]. None of the participants had any neurological, psychiatric, or other medical problems that are contraindicated for TMS [21]. Eye movement recording Binocular eye movements were recorded using an infrared tracker (Eyelink 1k; SR Research, Osgoode, Ontario, Canada) with a frequency of 500 Hz and a spatial resolution of 0.5 . Before the onset of each session, the eye tracker was calibrated with the standard 9-point Eyelink procedure. Before each trial, calibration was checked at fixation, and if the distance between the recorded position and the calibration was greater than 1, a new calibration was initiated. Experiment Builder software (SR Research, Osgoode, Ontario, Canada) allowed on-line monitoring of eye movements, presentation of visual stimuli, and triggering of the TMS train relative to the time of saccade detection. Eye-movement data were stored for off-line analysis. Instantaneous velocity and acceleration

were computed for each data sample and compared to a threshold (30 /s and 8000 /s2). Saccade onset was defined as two consecutive above-threshold samples for both criteria. Saccade offset was defined as the beginning of the next 20-ms period of belowthreshold samples.

Transcranial magnetic stimulation The TMS train (3 pulses, 100ms; for a similar procedure see ref. [22]) was delivered using a Magstim Rapid stimulator (Magstim, Whitland, Dyfed, U.K.) and a figure-of-eight coil (70mm). Prior to the experimental session, coil position was identified on each participant's scalp using the SofTaxic Navigator system (EMS, Bologna, Italy). This method has good localization accuracy (5 mm), with a level of precision close to that obtained using individual MRI [23]. Such a procedure has been successfully used in previous single-pulse TMS and rTMS studies investigating the functional role of parietal and frontal regions of the brain in various tasks [24e26]. First, skull landmarks (nasion, inion, and two preauricular points) and about 60 points providing a uniform representation of the scalp were digitized by means of a Polaris Vicra Optical Tracking System (NDI, Canada). Coordinates in Talairach space [27] were automatically estimated by the SofTaxic Navigator from an MRI-constructed stereotaxic template. Then, we selected the scalp sites corresponding to PPC and FEF regions in the right hemisphere. Scalp positions corresponding to PPC and FEF were identified by means of the SofTaxic Navigator system. The right PPC was targeted in the posterior part of the intraparietal sulcus (coordinates: x ¼ 27, y ¼ 84, z ¼ 48), and the right FEF region was targeted in the vicinity of the pre-central sulcus and the most dorsal part of the superior frontal sulcus (x ¼ 31, y ¼ -2, z ¼ 47). Effective stimulation of each of these sites has been previously shown in a range of studies [22,28e33]. The PPC and FEF scalp sites were marked on the cap with a pen. Then, the neuronavigation system was used to estimate the projections of the scalp sites on the brain surface. Stimulation of PPC or FEF was carried out by placing the coil tangentially over the marked scalp sites. We further controlled for accurate localization of the right FEF using an additional two-step procedure. First, we checked that the estimated cortical site was about 2e3 cm anterior to the motor hot spot of the primary motor contralateral hand area, located by the measurement of the individual resting motor threshold (see below) [34e40]. Second, we checked that stimulating the estimated FEF site at an intensity equivalent to that used during the experiment (see below) did not elicit any visible motor twitches in the resting contralateral hand [32]. Sham stimulation was performed by placing the coil over the vertex, localized as the point lying midway between the intertragal notches of the ears and midway between the inion and nasion. The coil was held by an articulated coil stand (Magstim) and its position with respect to the target sites on the standard reconstructed brain was continuously monitored during the experiment. Because motor threshold does not adequately represent the excitability threshold of non-motor cortical areas [41e44], we fixed the stimulation intensity at 60% of the maximum stimulator output (2T) and held it constant across subjects and across experiments (for similar procedure see refs 31,45,46,47). This intensity is approximately equivalent to the average intensity reached to establish the participants' resting motor threshold (group averaged rMT ¼ 58%, range ¼ 45%e74%). Individual rMT was measured as the lowest intensity for which TMS applied over the right primary motor cortex induces a visible twitch in the resting contralateral hand in 5 of 10 trials [48,49].

Please cite this article in press as: Collins T, Jacquet PO, TMS over posterior parietal cortex disrupts trans-saccadic visual stability, Brain Stimulation (2017), https://doi.org/10.1016/j.brs.2017.11.019

T. Collins, P.O. Jacquet / Brain Stimulation xxx (2017) 1e10

Experimental procedures Online TMS sessions were run in a dark room. Subjects sat 57cm away from a 2400 CRT monitor with a resolution of 1920  1080 pixels and a refresh rate of 85 Hz. Their head was stabilized by chin, cheekbone and forehead rests. Stimuli were 0.5 -diameter black dots on a light grey background (16.5 cd/m2). Double-step task Subjects were required to fixate, after eyetracker calibration, a black dot at the center of the screen and press a button, which initiated the fixation check. If successful, the trial began. The fixation dot remained present and 500ms later, the first target appeared for 250ms, followed by a 250ms blank and then the second target for 250ms. After a jitter of a random 250-500ms duration, the fixation dot disappeared, instructing the subject to make two memory-guided saccades to the remembered locations of the targets, in the order they had been presented (Fig. 1). The first target could randomly appear 10 left or 10 right of the fixation dot. The second target could randomly appear 8 above or below the first target (Fig. 1A). The task was divided into three separate blocks, corresponding to a particular stimulation condition (FEF, PPC and vertex), the order of which was counterbalanced between participants. Each block was composed of 148 trials (74 trials  2 saccade directions) with TMS, interleaved with 48 trials without TMS (24 trials  2 saccade directions). In TMS trials, the stimulation train was triggered as soon as the recorded eye position was outside a 4  4 window around the target. In control (no TMS) trials, feedback was displayed on the screen prompting subjects to be more accurate whenever the landing position of the first saccade was outside of an area within a 4 -diameter circle centered on the first target. In addition, if a saccade was detected before the go-signal, a warning appeared on the screen prompting subjects to maintain fixation.

3

Trials with such warnings were cancelled and reintroduced into the trial loop. This method was used to encourage subjects to make accurate saccades without increasing the number of TMS trials. In the double-step task trials were separated by an inter-trial interval (ITI) of 5 s. The ITIs of the two tasks led to an inter-pulse interval of approximately 10 s, hence minimizing a putative contamination of subsequent trials [50e53]. Finally, subjects were familiarized with the tasks by performing a set of 10 control trials, and with TMS stimulation by performing 5 trials with TMS (for each stimulation condition). We reconstructed the exact timing of TMS stimulation, based on time stamps in the eye movement data files (relevant transmission delays are on the order of