Human thalamus contributes to perceptual stability across eye movements Florian Ostendorfa,1, Daniela Liebermanna,b, and Christoph J. Plonera a

Department of Neurology and bCenter for Stroke Research, Charité-Universitätsmedizin Berlin, 10117 Berlin, Germany

Edited by Robert H. Wurtz, National Eye Institute, National Institutes of Health, Bethesda, MD, and approved December 3, 2009 (received for review September 18, 2009)

| corollary discharge | efference copy | extraretinal

T

o preserve visual stability across eye movements, the brain must distinguish self-induced image displacements from movements in the outside world. Visual reafferent information and optic flow patterns may provide important clues for this distinction (1). Sole reliance on visual reafference may not suffice, however, to fully disentangle self-induced from external visual changes: Motion perception is strongly suppressed during saccadic eye movements (2, 3), restricting the cues provided by optic flow. Transsaccadic memory of a visual scene seems to be surprisingly sparse, limiting the amount of relational information retained across fixations (4). Finally, many everyday situations with changing and unstable surroundings render relative positional information unreliable, e.g., when navigating through traffic. The brain may use extraretinal information to compensate for self-induced changes of the visual reafference (5). In more recent conceptualizations, an efference copy or corollary discharge of the oculomotor command may be fed into an internal forward model (6, 7) to generate a prediction of the visual reafference after a saccadic eye movement. Mismatches between predicted and actual visual reafference could then correctly be attributed to changes in the visual environment. To what degree extraretinal information shapes our visual experience across eye movements is still a matter of debate. Psychophysical studies demonstrated that normal human subjects frequently fail to detect visual displacements that happen within a saccadic eye movement (saccadic suppression of displacement) (8–10). However, sensitivity for intrasaccadic displacement detection can be surprisingly high under simple modifications of the original task (9, 11). Accurate and precise extraretinal monitoring information thus in principle is available to the human visual system, but how this information is integrated (10, 12) and what perceptual alterations might arise from deficient corollary discharge is still unclear. Recent neurophysiological studies in primates have identified a candidate pathway for a corollary discharge signal that ascends from

www.pnas.org/cgi/doi/10.1073/pnas.0910742107

Results N.P. is a 38-year-old right-handed man who had a thalamic stroke 10 months before testing. While mountaineering, he noticed a mild weakness of his left arm and leg that fully recovered within 30 min. In addition, he noticed memory problems that dissipated within the following days. On transferral to our department, the patient’s neurological examination was normal. Neuropsychological testing revealed normal intelligence and no impairment in visuo-spatial perception, memory and executive functions (Experimental Procedures). However, magnetic resonance imaging with subsequent lesion reconstruction revealed a highly focal and selective lesion of the right thalamus, confined to intralaminar nuclei and lateral portions of the right mediodorsal nucleus (16) (Fig. 1). The affected portions of central thalamus closely correspond to homologous portions of primate thalamus in which saccade- and eye positionrelated signals have been recorded previously (13, 17). Although single lesions affecting the thalamus are occasionally encountered in larger cohorts of stroke patients (18), oculomotor abnormalities and other neurological and cognitive deficits are observed in the vast majority of these patients (14, 19). Screening of the medical records and imaging databases in our department (>2,000 stroke patients/year) yielded no further subjects with unilateral lesions of similar selectivity in our thalamic region of interest. In this report, we therefore focus on patient N.P., who sustained a small and unilateral lesion just within this region and performed normally on neurological, neuropsychological and oculomotor examination. In a first step, we assessed N.P.’s performance in a double-step task. In this task, subjects perform two sequential saccades to the locations of two flashed targets (20). In a critical condition

Author contributions: F.O. and C.J.P. designed research; F.O. and D.L. performed research; F.O., D.L., and C.J.P. analyzed data; and F.O., D.L., and C.J.P. wrote the paper. The authors declare no conflict of interest. This article is a PNAS Direct Submission. To whom correspondence should be addressed. E-mail: fl[email protected].

1

This article contains supporting information online at www.pnas.org/cgi/content/full/ 0910742107/DCSupplemental.

PNAS | January 19, 2010 | vol. 107 | no. 3 | 1229–1234

NEUROSCIENCE

saccade

the midbrain to the frontal cortex, relayed by the mediodorsal thalamus (13). Pharmacological inactivation studies in nonhuman primates (13) and lesion studies in humans (14) point to an involvement of this pathway in the planning of eye movement sequences. Inactivation studies furthermore demonstrated that predictive remapping of visual responses in the frontal eye fields relies on the integrity of these pathways (15). The latter finding establishes a potential link between transthalamic monitoring signals and perceptual processing in nonhuman primates. However, a direct role of transthalamic pathways in the preservation of perceptual stability across eye movements has not been demonstrated in human subjects so far. In the present study, we assessed the role of these pathways for perceptual stability in a patient with a small and selective lesion affecting the right central thalamus. This patient provided the unique opportunity to study the putative perceptual consequences of impaired corollary discharge function in a human subject with uncompromised visual and oculomotor pathways.

PSYCHOLOGICAL AND COGNITIVE SCIENCES

We continuously move our eyes when we inspect a visual scene. Although this leads to a rapid succession of discontinuous and fragmented retinal snapshots, we perceive the world as stable and coherent. Neural mechanisms underlying visual stability may depend on internal monitoring of planned or ongoing eye movements. In the macaque brain, a pathway for the transmission of such signals has been identified that is relayed by central thalamic nuclei. Here, we studied a possible role of this pathway for perceptual stability in a patient with a selective lesion affecting homologous regions of the human thalamus. Compared with controls, the patient exhibited a unilateral deficit in monitoring his eye movements. This deficit was manifest by a systematic inaccuracy both in successive eye movements and in judging the locations of visual stimuli. In addition, perceptual consequences of oculomotor targeting errors were erroneously attributed to external stimulus changes. These findings show that the human brain draws on transthalamic monitoring signals to bridge the perceptual discontinuities generated by our eye movements.

Fig. 1. Lesion reconstruction of patient N.P. Structural MRI of N.P. shows a selective and focal ischemic lesion within the right central thalamus (RH, right hemisphere). Horizontal (A) and coronal (C) reconstructions at the level of maximal lesion extent were registered to digital drawings of thalamic nuclei, adapted from plates of an established atlas of the human thalamus (16) and fitted to the thalamic contours of N.P. (B and D). Contours of intralaminar nuclei are drawn in orange.

(EXTRARETINAL), the targets disappeared before the eyes started to move. Correct execution of the second saccade then depends on an internal estimate of the first saccade metrics (13, 14, 20). We compared the oculomotor performance in this condition with a control condition (VISUAL), in which no internal monitoring information was required for the correct execution of the saccade sequence (SI Experimental Procedures and Fig. S1). Similar to findings of a recent inactivation study in primates (13), N.P. exhibited a lateralized deficit for saccade sequences whenever an eye movement had to rely on the internal monitoring of a preceding saccadic eye movement. This deficit manifested as a systematic outward deviation of second saccade vectors for rightward-directed first saccades (i.e., directed ipsilateral to lesion side; Fig. 2), consistent with an internal monitoring signal that systematically underestimates corresponding saccade amplitudes (i.e., a hypometric corollary discharge signal; SI and Fig. S2). The unique selectivity of N.P.’s lesion then led us to directly assess a possible role of this thalamic region for stable space perception across saccades. In the task used, subjects followed a sudden stimulus jump with a saccadic eye movement. At saccade onset, the target was displaced to a variable degree, and subjects were instructed to report the apparent intrasaccadic jump direction after the saccade (STEP condition; Fig. 3A). We quantified the individual magnitude of displacement detection by determining the absolute jump size required to obtain a threshold level of 75% correct responses. When patient N.P. was tested in the STEP condition, a target displacement of roughly 1° was needed to achieve this threshold level (thresholds, 0.73° and 0.93° for left- and rightward saccades, respectively; Figs. 3B and 4). These detection thresholds lay within the performance range of a group of eight agematched male control subjects [thresholds (±SD), 1.18° (±0.33°) and 1.2° (±0.28°), respectively; Fig. 4B]. Statistical analysis confirmed that N.P.’s perceptual performance in the STEP condition was not significantly different from the control group for both saccade directions (two-tailed modified t test (21), displacement thresholds for leftward saccades, t = −1.3, P = 0.24; rightward saccades, t = −0.88, P = 0.41). 1230 | www.pnas.org/cgi/doi/10.1073/pnas.0910742107

Fig. 2. (A) Spatial layout of double-step task. Stimulus sequences started with fixation at screen center (F). The first target (T1) was presented horizontally at 6° or 10°, left or right from screen center. The second target (T2) was presented at 6° right or left and 6° above or below the screen center. Lines illustrate two of eight possible saccade sequences. (B) Double step performance of patient N.P. Mean start and end positions (±SEM) are shown for saccades sequences in the EXTRARETINAL and VISUAL condition (red and blue circles, respectively). For saccade sequences with a second downward saccade, vertical position information was inverted. Data are shown separately for the four resulting sequence configurations for leftward and rightward first saccades. P values refer to significant differences in second saccade endpoints between EXTRARETINAL and VISUAL condition (MannWhitney rank sum test); n.s.d., not significantly different. From a total of 337 valid trials entering analysis, 43/48 trials (EXTRARETINAL/VISUAL) contributed to the configuration with 6° leftward amplitude and 45/42 trials to the configuration with 6° rightward first saccade, 36/43 and 36/44 trials to configurations with 10° leftward and rightward first saccades, respectively.

For a more critical test of corollary discharge integrity, we used a simple modification of the original task in which a short gap of 250ms duration was introduced before the target reappeared (GAP condition; Fig. 3A). Our rationale was that target blanking may indicate a less reliable visual environment and may thus strengthen the weight that the brain attributes to internal monitoring signals (9, 10). In normal subjects, this procedure has been shown to make intrasaccadic target jumps more visible (9). In accordance with previous findings (9), all control subjects showed considerable improvement in detection thresholds in the GAP condition for both saccade directions [mean threshold reduction (±SD), 55% (± 22%); average threshold, 0.51° (±0.22°); Fig. 4B]. Thus, reliable internal information to predict the perceptual outcome of a saccadic eye movement was in principle available in normal subjects. A similar improvement in displacement detection was observed for leftward saccades in patient N.P. (threshold reduction, 27%; threshold, 0.53°; Figs. 3C and 4) and this threshold was not significantly different from the control group (two-tailed modified t test, t = −0.06, P = 0.96). By contrast, a dramatically different Ostendorf et al.

pattern emerged for N.P.’s rightward saccades with performance now deteriorating in the GAP condition (threshold increase, 58%; threshold, 1.47°; Figs. 3C and 4). This threshold was significantly different from the control group (two-tailed modified t test, t = 3.63, P = 0.009). We also examined whether the change in perceptual performance between STEP and GAP conditions was significantly different in patient N.P. compared with our control group. For leftward saccades, the performance change in N.P. was statistically indistinguishable from the change observed in the control group [two-tailed revised standardized difference test (22), t = 0.94, df = 7, P = 0.38]. For rightward saccades, however, a significantly different change in perceptual performance was confirmed (t = 2.73, df = 7, P = 0.03). A detailed analysis of N.P.’s eye movements showed that these findings cannot be explained by differences in the metrics of primary or corrective saccades (SI Text and Figs. S3 and S4). The observed deficit in intrasaccadic displacement detection for rightward saccades in the GAP task was mainly caused by a Ostendorf et al.

PNAS | January 19, 2010 | vol. 107 | no. 3 | 1231

NEUROSCIENCE

shift of the psychometric function in the forward direction (Fig. 3C, Right). This indicates that N.P. had a systematic tendency to report a backward jump when the target was actually stationary or even displaced in the forward direction. Interestingly, this pattern of results could have been predicted from the results in the double-step task: With a systematic underestimation of saccade amplitudes due to hypometric corollary discharge, N.P. should expect his saccades to land shorter of the target than they actually do. With a stationary target, N.P. will thus on average expect a larger perceptual forward error after saccade execution than actually experienced. Consequently, the target has to be shifted in the forward direction to achieve the impression of position constancy, and still larger forward jumps are required until N.P. attributes a forward displacement to a real target jump rather than to self-induced targeting errors. A calculation of the size of corollary discharge impairment for rightward saccades yielded similar quantitative estimates in the oculomotor and the perceptual detection task. Mean horizontal endpoint shift of second saccade endpoints in the double-step task was 1.44° (0.8° and 2.1° for 6° and 10° first saccade amplitudes, respectively) compared with a shift of the psychometric function in the perceptual detection task of 1.47° (for a mean saccade amplitude of 14°). These measures would correspond to a relative internal underestimation of actual saccade amplitude by on average 17% in the double-step task and 11% in the perceptual detection task. The similarity of these measures may indicate a single common underlying corollary discharge impairment assessed independently for oculomotor planning and perceptual reports. However, it should be kept in mind that these measures might reflect only an approximate estimate of corollary discharge inaccuracy, as other factors will likely influence the magnitude of the deficit (e.g., different experimental paradigms and analyses employed for the oculomotor and perceptual report data, a possible retinotopy/saccadotopy of the corollary discharge deficit).

PSYCHOLOGICAL AND COGNITIVE SCIENCES

Fig. 3. (A) Schematic of target displacement task. Trials started with a cue that jumped to the other screen side (black lines). Subjects followed with a saccadic eye movement (exemplary eye traces, gray lines). The eye movement triggered a saccade-contingent stimulus change: In the STEP condition, the target was displaced to a variable degree. In the GAP condition, a blank interval (250 ms) was introduced before the displaced target was shown again. Note the occurrence of secondary saccades that correct for the artificially induced targeting error of primary saccades. (B and C) Psychometric functions of patient N.P. Proportion of trials in which patient N.P. reported an apparent stimulus jump in saccade direction (forward), plotted against relative displacement levels. Negative values refer to target displacements against saccade direction. Vertical dashed line denotes zero displacement. Circle sizes represent the number of trials for a given target jump. Cumulative gaussians were fitted to perceptual response data separately for leftward and rightward saccades in the STEP (B, total of 103/95 valid trials for leftward/rightward saccades, respectively) and GAP (C; total of 160/184 valid trials, respectively) condition. Gray curves show averaged psychometric function of control group (with dashed lines showing ±SD).

Fig. 4. Perceptual displacement thresholds of patient N.P. (A) Psychometric functions of N.P.’s perceptual report, converted into percent correct vs. absolute displacement. Thresholds were calculated as the absolute displacement needed to obtain 75% correct responses (dashed vertical lines). Data are shown separately for leftward (contralateral) and rightward (ipsilateral) saccades in the STEP (blue curve) and GAP task (red curve). (B) Lower panels show resulting thresholds in N.P. and the control group. Gray bars show mean thresholds for control subjects (±SD); circles indicate thresholds of patient N.P.

The results so far suggest that the internal monitoring signals N.P. recruits for both movement planning and perceptual decisions systematically underestimate the actual movement amplitude of rightward saccades (i.e., the ipsilateral saccade direction with respect to lesion side). Deficient corollary discharge transmission might, however, lead not only to a systematically distorted but also to an increasingly random internal estimate of oculomotor actions. To investigate this possibility, we took advantage of the limited precision of oculomotor actions. Saccadic eye movements result from a cascade of neural processing steps involving sensorimotor integration, oculomotor planning, and final saccade execution. Each of these steps is inherently noisy and appears to contribute to a variable targeting error (23). In the normal brain, corollary discharge signals may inform perceptual decision stages about actual saccade targeting errors. Targeting errors from noise originating in processing steps upstream the level at which a putative corollary discharge signal is fed back should therefore be taken into account. The perceptual report should then be largely independent from the associated oculomotor targeting error (24, 25). For analysis, we sorted individual data in the GAP condition according to saccade targeting error and quantized them into bins of equal sample size (24). Then we correlated the binned perceptual report data with corresponding quantized saccade targeting errors. No significant correlation between quantized targeting error and corresponding average perceptual reports emerged in our control group (Spearman rho correlation, all individual P values >0.15; pooled group data, leftward saccades, r = 0.57, P = 0.14; rightward saccades, r = 0.44, P = 0.28; Fig. 5A). In agreement with recent psychophysical findings (24, 25), healthy individuals are apparently able to take the predominant part of their oculomotor imprecision into account for the evaluation of transsaccadic visual stability. The same held true for leftward saccades in patient N.P. (r = −0.2, P = 0.63; Fig. 5B Left). With

Fig. 5. Perceptual report of the control group (A) and patient N.P. (B), relative to corresponding saccade amplitude error in the GAP condition. The proportion of trials with forward reports is plotted against the average saccade targeting error for eight bins of equal sample size. Data for leftward (contralateral) and rightward (ipsilateral) saccades are shown separately. Positive targeting errors refer to hypometric primary saccades. Black squares in A show means of control group for a given bin with error bars indicating ± 95% CI. Insets note correlation coefficients and P values (Spearman rank correlation). Dashed lines denote chance response level.

1232 | www.pnas.org/cgi/doi/10.1073/pnas.0910742107

degraded internal monitoring, however, actual oculomotor targeting errors should increasingly contaminate perceptual reports. Indeed, a different picture emerged for N.P.’s rightward saccades: Here, this analysis yielded a highly significant correlation between targeting error and perceptual report (r = 0.95, P = 0.001; Fig. 5B Right). Targeting errors of rightward saccades thus almost perfectly accounted for the variance of N.P.’s corresponding perceptual report. This result points to a more pronounced impairment than the initial analysis of perceptual performance suggested. Success of transsaccadic space integration in the patient is at the mercy of the actual precision of his oculomotor output. Discussion This case study shows that a central thalamic lesion can lead to a specific and lateralized impairment in transsaccadic visual integration. Our patient suffered both from a systematic distortion of transsaccadic space integration and an additional incapability to keep track of oculomotor targeting errors: Perceptual consequences of oculomotor targeting errors were largely misattributed to external stimulus changes. This perceptual deficit can most parsimoniously be explained by the reliance on hypometric and more variable corollary discharge signals for rightward saccades. The systematic distortion of transsaccadic space integration matched the pattern of oculomotor planning deficits observed in our patient and in nonhuman primates after transient pharmacological inactivation of homologous thalamic regions (13). Our study thus significantly extends previous findings in nonhuman primates (13) and in patients with thalamic lesions (14) by showing that deficient internal monitoring signals indeed affect perceptual stability across eye movements. For both movement planning and perceptual decisions, corollary discharge-dependent performance in N.P. was predominantly impaired for rightward saccadic eye movements, i.e., for the saccade direction ipsilateral to lesion side. At first hand, this seems to be at odds with findings of an inactivation study that demonstrated monitoring deficits in a double-step task for saccades directed contralateral to the inactivated thalamus (13). However, sampling in this study was deliberately restricted to neurons driven from the ipsilateral superior colliculus and projecting to the ipsilateral frontal eye fields. The behavioral outcome could therefore be expected, as each superior colliculus codes for contralateral space (26). A recent study demonstrated that the frontal eye field receives information from both superior colliculi and hence the entire visual field (27). Projections from the contralateral superior colliculus may hereby cross at tectal and thalamic levels (27, 28). Indeed, electrophysiological work in primates indicates that corollary discharge-related activity at the thalamic level represents both ipsi- and contralateral saccades and eye positions (17, 29). Patient studies using a double-step task noted contra- and bilateral impairments with thalamic lesions but also strictly ipsilateral deficits in several cases (14, 30). The observed lateralization of N.P.’s deficit points to a predominant affection of thalamic portions conveying ipsilateral corollary discharge information and comprising either bilateral projections from other brainstem structures (17) or postcommissural fibers from the contralateral superior colliculus (27, 28). Why did our patient not consciously experience a breakdown of visual stability in everyday life? Although proprioceptive information about eye position seems to play a negligible role for the maintenance of visual stability (31), a number of alternative corollary discharge pathways may help to partially compensate for the affected transthalamic pathway (32). Visual stability plays a fundamental role for any kind of visually guided behavior and should therefore be maintained by robust and redundant mechanisms. This is in keeping with the observation that only large and bilateral brain pathology may occasionally lead to the subjective experience of compromised visual stability (33). We hypothesize that residual internal monitoring in N.P. suffices to Ostendorf et al.

Experimental Procedures Subjects. Eight healthy right-handed male subjects served as controls (mean age, 39.5 years; range, 31–54 years). The subjects had no history of neurological or psychiatric disorders, and all but one were naive with respect to the purpose of the study. Informed consent was obtained from patient N.P. and the control subjects before participation in the study, which was approved by the local Ethical Committee. Neuropsychological Testing in Patient N.P. Neuropsychological testing of patient N.P. revealed normal intelligence (as assessed with subtest no. 3 of LPS and the MWT-B, German equivalents to Raven’s Progressive Matrices and the National Adult Reading Test, respectively) and no impairment in visuo-spatial perception, memory, and executive functions (as assessed with forward and backward Digit and Block span, immediate and delayed recall of the Rey-Osterrieth figure, Wisconsin Card-Sorting-Test and Stroop-Test, respectively). Imaging and Lesion Reconstruction. Morphological imaging was performed on a clinical whole-body scanner (1.5 T. Magnetom Vision, Siemens) using a standard head coil. To screen for extra- and intrathalamic lesions at the time of testing, axial images of the whole brain and coronal images of the thalamic region were acquired using a turbo inversion recovery magnitude sequence. No further lesion was detected, confirming the selective nature of the described thalamic lesion. For lesion reconstruction, a three-dimensional data set at high resolution was acquired using a magnetization prepared rapid acquisition gradient-echo imaging sequence (MPRAGE, isotropic resolution 1 mm). The

1. Gibson JJ (1966) The Senses Considered as Perceptual Systems (Houghton Mifflin, Boston). 2. Burr DC, Morrone MC, Ross J (1994) Selective suppression of the magnocellular visual pathway during saccadic eye movements. Nature 371:511–513.

Ostendorf et al.

midsagittal plane and a horizontal plane perpendicular to midsagittal plane and running through the anterior and posterior commissures (AC-PC plane) were used as reference planes. Axial and coronal reconstructions from the highresolution data set were calculated with respect to these reference planes. These reconstructions were then registered to digital line drawings of thalamic nuclei, drawn after an established atlas of the human thalamus (16).

Intrasaccadic Target Displacement. A fixation cross (extent, 0.5°) was presented at 6° or 8° lateral from screen center. After a variable foreperiod (1,600–2,400 ms), the fixation cross was switched off and a target cue (diameter, 0.5°) was simultaneously presented at the other side of the screen at 6° or 8° eccentricity, respectively. Variable start and target positions were chosen to make the target position less predictable and to prevent subjects from performing stereotyped saccades. In the main experiment, the target for a saccadic eye movement was switched off during saccade execution and reappeared either directly (STEP condition) or after a temporal gap of 250 ms (GAP condition) at an unpredictable position. Target displacement for a given trial was adapted by three independent, randomly interleaved staircases with a constant step size of 3° in the STEP task (GAP task, 1.5°). Specifically, when the subject indicated a target displacement to the left for a given displacement level, the next probe for a given staircase would be shifted by 3° (GAP task, 1.5°) to the right. Staircases started at a displacement level of 7° (GAP task, 3.5°) right- and leftward and 0° (no displacement) with respect to initial target position. Interleaved displacement levels for the three staircases enabled sampling the point of subjective target constancy with a resolution of 1° (GAP task, 0.5°) while collecting a sufficient number of trials at higher confidence levels. In both conditions subjects reported the apparent jump direction by pressing one of two manual response buttons. Subjects were instructed to randomly press one of the buttons if they were guessing. Response registration was limited to maximally 5 s and the target was switched off when a button press was recorded or maximum response time had elapsed. The screen was then blanked for 1,600 ms and a next trial started. Saccade direction was fixed within five to six blocks of 24 trials each. Eye movement data were low-pass filtered, visualized and analyzed in Matlab (Mathworks) by using the ILAB toolbox (37) and self-written routines. Saccade onset and offset were determined by a fixed velocity criterion (threshold, 30°/s). Saccade start and end positions were determined as fixation periods preceding saccade onset and following saccade end, respectively (with fixations defined as sets of consecutive eye position samples with less than 0.5° dispersion for at least 80 ms). We ensured that intrasaccadic target displacements could be realized in the first half of the saccadic eye movement [mean delay after saccade onset (±SD), 19 ms (± 4 ms)]. Cumulative gaussians were fitted to the perceptual response data in Matlab by using psignifit, a toolbox that implements the maximum-likelihood method described by Wichmann and Hill (38). Uncertainty and bias of subjective displacement reports were described by slope and threshold estimates of the fitted psychometric function (with steeper psychometric curves yielding higher slope values). For easier comparison between conditions and subjects, we converted psychometric functions to percent correct values and calculated the resulting psychometric functions for absolute displacement values. This enabled us to determine a displacement threshold, i.e., the displacement necessary to obtain correct responses in 75% of trials for a given condition and subject. ACKNOWLEDGMENTS. We are very grateful to N.P. We thank two anonymous reviewers for helpful comments on a previous version of the manuscript. This work was supported by German Federal Ministry of Education and Research Grant 01GW0653 (Visuospatial-Cognition).

3. Thiele A, Henning P, Kubischik M, Hoffmann KP (2002) Neural mechanisms of saccadic suppression. Science 295:2460–2462. 4. Irwin DE (1991) Information integration across saccadic eye movements. Cognit Psychol 23:420–456.

PNAS | January 19, 2010 | vol. 107 | no. 3 | 1233

NEUROSCIENCE

General Experimental Set-Up. Subjects sat at a viewing distance of 50 cm in front of a 22-in. CRT-monitor (refresh rate, 110 Hz) with their heads stabilized by a head- and chinrest. Eye movements were recorded with high-speed video-oculography (Sensomotoric Instruments) at a sampling rate of 500 Hz. Experiments were carried out in an otherwise darkened room. Subjects completed the experiments in multiple test sessions on different days. All stimuli were white (luminance, 56.5 Cd/m2) and presented on a homogenous gray background (luminance, 13.1 Cd/m2). Relatively high background luminance levels were chosen to exclude any spurious effects of phosphor persistence. Previous work demonstrated that visible screen borders should not influence localization with our stimulus configuration (36).

PSYCHOLOGICAL AND COGNITIVE SCIENCES

maintain visual stability under the default hypothesis that translations of the whole visual scene are unlikely and can thus be discarded within a certain tolerance range (9, 12). Such an internal presupposition of a stationary world is consistent with the general phenomenon of saccadic suppression of displacement in normal subjects (9, 10) and would explain N.P.’s inconspicuous perceptual performance in our STEP condition, mimicking a reductionist stationary world. Temporal contiguity of visual stimuli across saccades and constancy of relative positional information in a visual scene may thus mainly account for visual stability in most everyday situations (1, 9). Even a distorted and coarse internal monitoring signal might, under these circumstances, convey sufficient temporal and spatial information to complement the evaluation of visual reafferent information. Our findings in the GAP condition nevertheless show that impaired corollary discharge transmission comes at a perceptual cost: Whenever the brain starts to question the prior assumption of a stationary world, visual space perception across eye movements is compromised and affected by oculomotor noise. Transthalamic monitoring signals may thus significantly contribute to the correct attribution of self-induced vs. externally imposed changes in the continuous flow of our sensory experiences. This lends direct experimental support to the longstanding hypothesis of an important contribution of corollary discharge signals to the perceptual integration of space across eye movements (5). One might speculate that under specific environmental circumstances (e.g., intermittent illumination of the visual scene, a rapidly changing and unstable environment), deficient transsaccadic space integration as encountered in our patient for rightward saccades might occasionally lead to perceptible jumps of the outside world and thus might ultimately disturb the subjective maintenance of visual stability. On a general level, the present paradigm highlights the role of transthalamic processing loops for the successful self-monitoring of an individual interacting with the environment. Central aspects of our self-conception may build upon the integration of such corollary discharge-transmitting loops (34) and their disturbed functioning might contribute to the symptomatology in devastating diseases such as schizophrenia (35).

5. von Helmholtz H (1867) Handbuch der Physiologischen Optik (Leopold Voss, Leipzig). 6. Wolpert DM, Miall RC (1996) Forward models for physiological motor control. Neural Netw 9:1265–1279. 7. Crapse TB, Sommer MA (2008) The frontal eye field as a prediction map. Prog Brain Res 171:383–390. 8. Bridgeman B, Hendry D, Stark L (1975) Failure to detect displacement of the visual world during saccadic eye movements. Vision Res 15:719–722. 9. Deubel H, Schneider WX, Bridgeman B (1996) Postsaccadic target blanking prevents saccadic suppression of image displacement. Vision Res 36:985–996. 10. Niemeier M, Crawford JD, Tweed DB (2003) Optimal transsaccadic integration explains distorted spatial perception. Nature 422:76–80. 11. Gysen V, De Graef P, Verfaillie K (2002) Detection of intrasaccadic displacements and depth rotations of moving objects. Vision Res 42:379–391. 12. MacKay DM (1972) Visual stability. Invest Ophthalmol 11:518–524. 13. Sommer MA, Wurtz RH (2002) A pathway in primate brain for internal monitoring of movements. Science 296:1480–1482. 14. Bellebaum C, Daum I, Koch B, Schwarz M, Hoffmann KP (2005) The role of the human thalamus in processing corollary discharge. Brain 128:1139–1154. 15. Sommer MA, Wurtz RH (2006) Influence of the thalamus on spatial visual processing in frontal cortex. Nature 444:374–377. 16. Morel SA (2007) Stereotactic Atlas of the Human Thalamus and Basal Ganglia (Informa Healthcare, New York). 17. Tanaka M (2007) Spatiotemporal properties of eye position signals in the primate central thalamus. Cereb Cortex 17:1504–1515. 18. Bogousslavsky J, Van Melle G, Regli F (1988) The Lausanne Stroke Registry: Analysis of 1,000 consecutive patients with first stroke. Stroke 19:1083–1092. 19. Schmahmann JD (2003) Vascular syndromes of the thalamus. Stroke 34:2264–2278. 20. Hallett PE, Lightstone AD (1976) Saccadic eye movements to flashed targets. Vision Res 16:107–114. 21. Crawford JR, Garthwaite PH (2002) Investigation of the single case in neuropsychology: Confidence limits on the abnormality of test scores and test score differences. Neuropsychologia 40:1196–1208. 22. Crawford JR, Garthwaite PH (2005) Testing for suspected impairments and dissociations in single-case studies in neuropsychology: Evaluation of alternatives using monte carlo simulations and revised tests for dissociations. Neuropsychology 19:318–331.

1234 | www.pnas.org/cgi/doi/10.1073/pnas.0910742107

23. van Beers RJ (2007) The sources of variability in saccadic eye movements. J Neurosci 27:8757–8770. 24. Collins T, Rolfs M, Deubel H, Cavanagh P (2009) Post-saccadic location judgments reveal remapping of saccade targets to non-foveal locations. J Vis 9(29):1–9. 25. Munuera J, Morel P, Duhamel JR, Deneve S (2009) Optimal sensorimotor control in eye movement sequences. J Neurosci 29:3026–3035. 26. Wurtz RH, Goldberg ME (1972) Activity of superior colliculus in behaving monkey. 3. Cells discharging before eye movements. J Neurophysiol 35:575–586. 27. Crapse TB, Sommer MA (2009) Frontal eye field neurons with spatial representations predicted by their subcortical input. J Neurosci 29:5308–5318. 28. Preuss TM, Goldman-Rakic PS (1987) Crossed corticothalamic and thalamocortical connections of macaque prefrontal cortex. J Comp Neurol 257:269–281. 29. Wyder MT, Massoglia DP, Stanford TR (2003) Quantitative assessment of the timing and tuning of visual-related, saccade-related, and delay period activity in primate central thalamus. J Neurophysiol 90:2029–2052. 30. Gaymard B, Rivaud S, Pierrot-Deseilligny C (1994) Impairment of extraretinal eye position signals after central thalamic lesions in humans. Exp Brain Res 102:1–9. 31. Guthrie BL, Porter JD, Sparks DL (1983) Corollary discharge provides accurate eye position information to the oculomotor system. Science 221:1193–1195. 32. Sommer MA, Wurtz RH (2008) Brain circuits for the internal monitoring of movements. Annu Rev Neurosci 31:317–338. 33. Haarmeier T, Thier P, Repnow M, Petersen D (1997) False perception of motion in a patient who cannot compensate for eye movements. Nature 389:849–852. 34. Gallagher I, I (2000) Philosophical conceptions of the self: Implications for cognitive science. Trends Cogn Sci 4:14–21. 35. Feinberg I (1978) Efference copy and corollary discharge: Implications for thinking and its disorders. Schizophr Bull 4:636–640. 36. Deubel H (2004) Localization of targets across saccades: Role of landmark objects. Vis Cogn 11:173–202. 37. Gitelman DR (2002) ILAB: A program for postexperimental eye movement analysis. Behav Res Methods Instrum Comput 34:605–612. 38. Wichmann FA, Hill NJ (2001) The psychometric function: I. Fitting, sampling, and goodness of fit. Percept Psychophys 63:1293–1313.

Ostendorf et al.