(The limits of) eye-tracking with iPads.

Applications for eye-tracking-particularly in the clinic-are limited by a reliance on dedicated hardware. Here we compare eye-tracking implemented on an Apple iPad Pro 11" (third generation)-using the device's infrared head-tracking and front-facing camera-with a Tobii 4c infrared eye-tracker. We estimated gaze location using both systems while 28 observers performed a variety of tasks. For estimating fixation, gaze position estimates from the iPad were less accurate and precise than the Tobii (mean absolute error of 3.2° ± 2.0° compared with 0.75° ± 0.43°), but fixation stability estimates were correlated across devices (r = 0.44, p < 0.05). For tasks eliciting saccades >1.5°, estimated saccade counts (r = 0.4-0.73, all p < 0.05) were moderately correlated across devices. For tasks eliciting saccades >8° we observed moderate correlations in estimated saccade speed and amplitude (r = 0.4-0.53, all p < 0.05). We did, however, note considerable variation in the vertical component of estimated smooth pursuit speed from the iPad and a catastrophic failure of tracking on the iPad in 5% to 20% of observers (depending on the test). Our findings sound a note of caution to researchers seeking to use iPads for eye-tracking and emphasize the need to properly examine their eye-tracking data to remove artifacts and outliers.

Evidence for an Association Between a pH-Dependent Potassium Channel, TWIK-1, and the Accuracy of Smooth Pursuit Eye Movements.

Purpose: Within the healthy population there is a large variation in the ability to perform smooth pursuit eye movements. Our purpose was to investigate the genetic and physiological bases for this variation. Methods: We carried out a whole-genome association study, recording smooth pursuit movements for 1040 healthy volunteers by infrared oculography. The primary phenotypic measure was root mean square error (RMSE) of eye position relative to target position. Secondary measures were pursuit gain, frequency of catch-up saccades, and frequency of anticipatory saccades. Ten percent of participants, chosen randomly, were tested twice, giving estimates of test-retest reliability. Results: No significant association was found with three genes previously identified as candidate genes for variation in smooth pursuit: DRD3, COMT, NRG1. A strong association (P = 3.55 × 10-11) was found between RMSE and chromosomal region 1q42.2. The most strongly associated marker (rs701232) lies in an intron of KCNK1, which encodes a two-pore-domain potassium ion channel TWIK-1 (or K2P1) that affects cell excitability. Each additional copy of the A allele decreased RMSE by 0.29 standard deviation. When a psychophysical test of visually perceived motion was used as a covariate in the regression analysis, the association with rs701232 did not weaken (P = 5.38 × 10-12). Conclusions: Variation in the sequence or the expression of the pH-dependent ion channel TWIK-1 is a likely source of variance in smooth pursuit. The variance associated with TWIK-1 appears not to arise from sensory mechanisms, because the use of a perceptual covariate left the association intact.

Evidence from comprehensive independent validation studies for smooth pursuit dysfunction as a sensorimotor biomarker for psychosis.

Smooth pursuit eye movements are considered a well-established and quantifiable biomarker of sensorimotor function in psychosis research. Identifying psychotic syndromes on an individual level based on neurobiological markers is limited by heterogeneity and requires comprehensive external validation to avoid overestimation of prediction models. Here, we studied quantifiable sensorimotor measures derived from smooth pursuit eye movements in a large sample of psychosis probands (N = 674) and healthy controls (N = 305) using multivariate pattern analysis. Balanced accuracies of 64% for the prediction of psychosis status are in line with recent results from other large heterogenous psychiatric samples. They are confirmed by external validation in independent large samples including probands with (1) psychosis (N = 727) versus healthy controls (N = 292), (2) psychotic (N = 49) and non-psychotic bipolar disorder (N = 36), and (3) non-psychotic affective disorders (N = 119) and psychosis (N = 51) yielding accuracies of 65%, 66% and 58%, respectively, albeit slightly different psychosis syndromes. Our findings make a significant contribution to the identification of biologically defined profiles of heterogeneous psychosis syndromes on an individual level underlining the impact of sensorimotor dysfunction in psychosis.

Exploring oculomotor functions in a pilot study with healthy controls: Insights from eye-tracking and fMRI.

Eye-tracking techniques have gained widespread application in various fields including research on the visual system, neurosciences, psychology, and human-computer interaction, with emerging clinical implications. In this preliminary phase of our study, we introduce a pilot test of innovative virtual reality technology designed for tracking head and eye movements among healthy individuals. This tool was developed to assess the presence of mild traumatic brain injury (mTBI), given the frequent association of oculomotor function deficits with such injuries. Alongside eye-tracking, we also integrated fMRI due to the complementary nature of these techniques, offering insights into both neural activation patterns and behavioural responses, thereby providing a comprehensive understanding of oculomotor function. We used fMRI with tasks evaluating oculomotor functions: Smooth Pursuit (SP), Saccades, Anti-Saccades, and Optokinetic Nystagmus (OKN). Prior to the scanning, the testing with a system of VR goggles with integrated eye and head tracking was used where subjects performed the same tasks as those used in fMRI. 31 healthy adult controls (HCs) were tested with the purpose of identifying brain regions associated with these tasks and collecting preliminary norms for later comparison with concussed subjects. HCs' fMRI results showed following peak activation regions: SP-cuneus, superior parietal lobule, paracentral lobule, inferior parietal lobule (IPL), cerebellartonsil (CT); Saccades-middle frontal gyrus (MFG), postcentral gyrus, medial frontal gyrus; Anti-saccades-precuneus, IPL, MFG; OKN-middle temporal gyrus, ACC, postcentral gyrus, MFG, CT. These results demonstrated brain regions associated with the performance on oculomotor tasks in healthy controls and most of the highlighted areas are corresponding with those affected in concussion. This suggests that the involvement of brain areas susceptible to mTBI in implementing oculomotor evaluation, taken together with commonly reported oculomotor difficulties post-concussion, may lead to finding objective biomarkers using eye-tracking tasks.

Smooth pursuit deficits impact dynamic visual acuity in macular degeneration.

SIGNIFICANCE: Prior studies with large, highly visible targets report low smooth pursuit gains in individuals with macular degeneration (MD). We show that lower gains persist even when observers are pursuing a target that requires discrimination at the acuity limit. This low gain causes retinal slip, potentially leading to motion blur and target disappearance in the scotoma, which further compromise the visibility of moving object. PURPOSE: In this study, we examine whether the characteristics of smooth pursuit (pursuit gain and placement of the fixational locus relative to the target) change when the task requires dynamic visual acuity. METHODS: Using the scanning laser ophthalmoscope, we recorded smooth pursuit eye movements in 10 eyes of 6 MD participants and 7 eyes of 4 age-matched controls in response to leftward- or rightward-moving annular targets (O) that briefly (300 milliseconds) changed to a Landolt C at one of several time points during the pursuit trial. Participants were asked to pursue the target and indicate the direction of the C opening. RESULTS: Individuals with MD had lower pursuit gains and fewer saccades during the C presentation than during the O, compared with their age-matched peers. Further, pursuit gain, but not the distance of the retinal pursuit locus from the target, predicted task performance in the MD group. CONCLUSIONS: Our findings suggest that compromised pursuit gain in MD participants likely further compromises their dynamic visual acuity and thus ability to view moving targets.

EEG Activity during Pursuit and Saccade Visual Strategies to Predict the Arrival Position of a Target.

BACKGROUND: In this study, we used electroencephalogram (EEG) to investigate the activity pattern of the cerebral cortex related to visual pursuit and saccade strategies to predict the arrival position of a visual target. In addition, we clarified the differences in the EEG of those who could predict the arrival position well using the saccade strategy compared to those who were not proficient. METHODS: Sixteen participants performed two tasks: the "Pursuit Strategy Task (PST)" and the "Saccade Strategy Task (SST)" while undergoing EEG. For the PST, the participants were instructed to follow the target with their eyes throughout its trajectory and indicate when it reached the final point. For the SST, the participants were instructed to shift their gaze to the end point of arrival once they had predicted it. RESULTS: Low beta EEG activity at the Oz, Cz, and CP2 electrodes was significantly higher during the SST than during the PST. In addition, low beta EEG activity at P7 electrode was significantly higher in the group showing a small position error (PE) than in the group showing a large PE at response. CONCLUSIONS: EEG activity at the Oz, Cz, and CP2 electrodes during the SST may reflect visuospatial attention to the moving target, the tracking of moving targets, and the focus on the final destination position. In addition, EEG activity at P7 electrode may more accurately detect the speed and direction of the moving target by the small PE group at response.

Interactions between Saccades and Smooth Pursuit Eye Movements in Marmosets.

Animals use a combination of eye movements to track moving objects. These different eye movements need to be coordinated for successful tracking, requiring interactions between the systems involved. Here, we study the interaction between the saccadic and smooth pursuit eye movement systems in marmosets. Using a single-target pursuit task, we show that saccades cause an enhancement in pursuit following a saccade. Using a two-target pursuit task, we show that this enhancement in pursuit is selective toward the motion of the target selected by the saccade, irrespective of any biases in pursuit prior to the saccade. These experiments highlight the similarities in the functioning of saccadic and smooth pursuit eye movement systems across primates.

Clinical value of video oculomotor evaluation in the differential diagnosis of multiple system atrophy and Parkinson's disease.

BACKGROUND: Multiple system atrophy (MSA) is a neurodegenerative disease that progresses rapidly and has a poor prognosis. This study aimed to assess the value of video oculomotor evaluation (VOE) in the differential diagnosis of MSA and Parkinson's disease (PD). METHODS: In total, 28 patients with MSA, 31 patients with PD, and 30 age- and sex-matched healthy controls (HC) were screened and included in this study. The evaluation consisted of a gaze-holding test, smooth pursuit eye movement (SPEM), random saccade, and optokinetic nystagmus (OKN). RESULTS: The MSA and PD groups had more abnormalities and decreased SPEM gain than the HC group (64.29%, 35.48%, 10%, p < .001). The SPEM gain in the MSA group was significantly lower than that in the PD group at specific frequencies. Patients with MSA and PD showed prolonged latencies in all saccade directions compared with those with HC. However, the two diseases had no significant differences in the saccade parameters. The OKN gain gradually decreased from the HC to the PD and the MSA groups (p < .05). Compared with the PD group, the gain in the MSA group was further decreased in the OKN test at 30°/s (Left, p = .010; Right p = .016). Receiver operating characteristic curves showed that the combination of oculomotor parameters with age and course of disease could aid in the differential diagnosis of patients with MSA and PD, with a sensitivity of 89.29% and a specificity of 70.97%. CONCLUSIONS: The combination of oculomotor parameters and clinical data may aid in the differential diagnosis of MSA and PD. Furthermore, VOE is vital in the identification of neurodegenerative diseases.

Gaze and attention: Mechanisms underlying the therapeutic effect of optokinetic stimulation in spatial neglect.

Left smooth pursuit eye movement training in response to large-field visual motion (optokinetic stimulation) has become a promising rehabilitation method in left spatial inattention or neglect. The mechanisms underlying the therapeutic effect, however, remain unknown. During optokinetic stimulation, there is an error in visual localisation ahead of the line of sight. This could indicate a change in the brain's estimate of one's own direction of gaze. We hypothesized that optokinetic stimulation changes the brain's estimate of gaze. Because this estimate is critical for coding the locus of attention in the visual space relative to the body and across sensory modalities, its change might underlie the change in spatial attention. Here, we report that in healthy participants optokinetic stimulation causes not only a directional bias in the proprioceptive signal from the extraocular muscles, but also a corresponding shift of the locus of attention. Both changes outlasted the period of stimulation. This result forms a step in investigating a causal link between the adaptation in the sensorimotor gaze signals and the recovery in spatial neglect.

Smooth pursuit inhibition reveals audiovisual enhancement of fast movement control.

The sudden onset of a visual object or event elicits an inhibition of eye movements at latencies approaching the minimum delay of visuomotor conductance in the brain. Typically, information presented via multiple sensory modalities, such as sound and vision, evokes stronger and more robust responses than unisensory information. Whether and how multisensory information affects ultra-short latency oculomotor inhibition is unknown. In two experiments, we investigate smooth pursuit and saccadic inhibition in response to multisensory distractors. Observers tracked a horizontally moving dot and were interrupted by an unpredictable visual, auditory, or audiovisual distractor. Distractors elicited a transient inhibition of pursuit eye velocity and catch-up saccade rate within ∼100 ms of their onset. Audiovisual distractors evoked stronger oculomotor inhibition than visual- or auditory-only distractors, indicating multisensory response enhancement. Multisensory response enhancement magnitudes were equal to the linear sum of responses to component stimuli. These results demonstrate that multisensory information affects eye movements even at ultra-short latencies, establishing a lower time boundary for multisensory-guided behavior. We conclude that oculomotor circuits must have privileged access to sensory information from multiple modalities, presumably via a fast, subcortical pathway.

Human estimates of descending objects' motion are more accurate than those of ascending objects regardless of gravity information.

Humans can accurately estimate and track object motion, even if it accelerates. Research shows that humans exhibit superior estimation and tracking performance for descending (falling) than ascending (rising) objects. Previous studies presented ascending and descending targets along the gravitational and body axes in an upright posture. Thus, it is unclear whether humans rely on congruent information between the direction of the target motion and gravity or the direction of the target motion and longitudinal body axes. Two experiments were conducted to explore these possibilities. In Experiment 1, participants estimated the arrival time at a goal for both upward and downward motion of targets along the longitudinal body axis in the upright (both axes of target motion and gravity congruent) and supine (both axes incongruent) postures. In Experiment 2, smooth pursuit eye movements were assessed while tracking both targets in the same postures. Arrival time estimation and smooth pursuit eye movement performance were consistently more accurate for downward target motion than for upward motion, irrespective of posture. These findings suggest that the visual experience of seeing an object moving along an observer's leg side in everyday life may influence the ability to accurately estimate and track the descending object's motion.

Different extrapolation of moving object locations in perception, smooth pursuit, and saccades.

The ability to accurately perceive and track moving objects is crucial for many everyday activities. In this study, we use a "double-drift stimulus" to explore the processing of visual motion signals that underlie perception, pursuit, and saccade responses to a moving object. Participants were presented with peripheral moving apertures filled with noise that either drifted orthogonally to the aperture's direction or had no net motion. Participants were asked to saccade to and track these targets with their gaze as soon as they appeared and then to report their direction. In the trials with internal motion, the target disappeared at saccade onset so that the first 100 ms of the postsaccadic pursuit response was driven uniquely by peripheral information gathered before saccade onset. This provided independent measures of perceptual, pursuit, and saccadic responses to the double-drift stimulus on a trial-by-trial basis. Our analysis revealed systematic differences between saccadic responses, on one hand, and perceptual and pursuit responses, on the other. These differences are unlikely to be caused by differences in the processing of motion signals because both saccades and pursuits seem to rely on shared target position and velocity information. We conclude that our results are instead due to a difference in how the processing mechanisms underlying perception, pursuit, and saccades combine motor signals with target position. These findings advance our understanding of the mechanisms underlying dissociation in visual processing between perception and eye movements.

Eye-hand coordination all the way: from discrete to continuous hand movements.

The differentiation between continuous and discrete actions is key for behavioral neuroscience. Although many studies have characterized eye-hand coordination during discrete (e.g., reaching) and continuous (e.g., pursuit tracking) actions, all these studies were conducted separately, using different setups and participants. In addition, how eye-hand coordination might operate at the frontier between discrete and continuous movements remains unexplored. Here we filled these gaps by means of a task that could elicit different movement dynamics. Twenty-eight participants were asked to simultaneously track with their eyes and a joystick a visual target that followed an unpredictable trajectory and whose position was updated at different rates (from 1.5 to 240 Hz). This procedure allowed us to examine actions ranging from discrete point-to-point movements (low refresh rate) to continuous pursuit (high refresh rate). For comparison, we also tested a manual tracking condition with the eyes fixed and a pure eye tracking condition (hand fixed). The results showed an abrupt transition between discrete and continuous hand movements around 3 Hz contrasting with a smooth trade-off between fixations and smooth pursuit. Nevertheless, hand and eye tracking accuracy remained strongly correlated, with each of these depending on whether the other effector was recruited. Moreover, gaze-cursor distance and lag were smaller when eye and hand performed the task conjointly than separately. Altogether, despite some dissimilarities in eye and hand dynamics when transitioning between discrete and continuous movements, our results emphasize that eye-hand coordination continues to smoothly operate and support the notion of synergies across eye movement types.NEW & NOTEWORTHY The differentiation between continuous and discrete actions is key for behavioral neuroscience. By using a visuomotor task in which we manipulate the target refresh rate to trigger different movement dynamics, we explored eye-hand coordination all the way from discrete to continuous actions. Despite abrupt changes in hand dynamics, eye-hand coordination continues to operate via a gradual trade-off between fixations and smooth pursuit, an observation confirming the notion of synergies across eye movement types.

TRACK-a new algorithm and open-source tool for the analysis of pursuit-tracking sensorimotor integration processes.

In daily life, sensorimotor integration processes are fundamental for many cognitive operations. The pursuit-tracking paradigm is an ecological and valid paradigm to examine sensorimotor integration processes in a more complex environment than many established tasks that assess simple motor responses. However, the analysis of pursuit-tracking performance is complicated, and parameters quantified to examine performance are sometimes ambiguous regarding their interpretation. We introduce an open-source algorithm (TRACK) to calculate a new tracking error metric, the spatial error, based on the identification of the intended target position for the respective cursor position. The identification is based on assigning cursor and target direction changes to each other as key events, based on the assumptions of similarity and proximity. By applying our algorithm to pursuit-tracking data, beyond replication of known effects such as learning or practice effects, we show a higher precision of the spatial tracking error, i.e., it fits our behavioral data better than the temporal tracking error and thus provides new insights and parameters for the investigation of pursuit-tracking behavior. Our work provides an important step towards fully utilizing the potential of pursuit-tracking tasks for research on sensorimotor integration processes.

Eye movements in Parkinson's disease: from neurophysiological mechanisms to diagnostic tools.

Movement disorders such as Parkinson's disease (PD) impact oculomotor function - the ability to move the eyes accurately and purposefully to serve a multitude of sensory, cognitive, and secondary motor tasks. Decades of neurophysiological research in monkeys and behavioral studies in humans have characterized the neural basis of healthy oculomotor control. This review links eye movement abnormalities in persons living with PD to the underlying neurophysiological mechanisms and pathways. Building on this foundation, we highlight recent progress in using eye movements to gauge symptom severity, assess treatment effects, and serve as potential precision biomarkers. We conclude that whereas eye movements provide insights into PD mechanisms, based on current evidence they appear to lack sufficient sensitivity and specificity to serve as a standalone diagnostic tool. Their full potential may be realized when combined with other disease indicators in big datasets.

Impairment of oculomotor functions in patients with early to advanced amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) can result into an incomplete locked in state (iLIS), in which communication depends on eye tracking computer devices. Oculomotor function impairments in ALS have been reported, but there is little research, particularly with respect to patients in iLIS. In the present study, we compared reflexive and executive oculomotor function by means of an eye tracking test battery between three groups: advanced ALS patients in iLIS (n = 22), patients in early to middle ALS stages (n = 44) and healthy subjects (n = 32). Patients with ALS showed significant deteriorations in oculomotor functions, with stronger impairments in iLIS. More specifically, ALS patients produced visually guided prosaccades with longer latencies and more frequent hypometria compared to healthy subjects. Longest latencies were obtained in iLIS patients, with a stronger prolongation for vertical than for horizontal prosaccades. ALS patients made more antisaccade errors and generated antisaccades with longer latencies. Smooth pursuit was also impaired in ALS. In the earlier ALS stages, bulbar onset patients presented stronger antisaccade and smooth pursuit deficits than spinal onset patients. Our findings reveal a relevant deterioration of important oculomotor functions in ALS, which increases in iLIS. It includes impairments of reflexive eye movements to loss of executive inhibitory control, indicating a progressing pathological involvement of prefrontal, midbrain and brainstem areas. The assessment of oculomotor functions may therefore provide clinically relevant bio- and progression marker, particularly in advanced ALS.

Bayesian approaches to smooth pursuit of random dot kinematograms: effects of varying RDK noise and the predictability of RDK direction.

Smooth pursuit eye movements respond on the basis of both immediate and anticipated target motion, where anticipations may be derived from either memory or perceptual cues. To study the combined influence of both immediate sensory motion and anticipation, subjects pursued clear or noisy random dot kinematograms (RDKs) whose mean directions were chosen from Gaussian distributions with SDs = 10° (narrow prior) or 45° (wide prior). Pursuit directions were consistent with Bayesian theory in that transitions over time from dependence on the prior to near total dependence on immediate sensory motion (likelihood) took longer with the noisier RDKs and with the narrower, more reliable, prior. Results were fit to Bayesian models in which parameters representing the variability of the likelihood either were or were not constrained to be the same for both priors. The unconstrained model provided a statistically better fit, with the influence of the prior in the constrained model smaller than predicted from strict reliability-based weighting of prior and likelihood. Factors that may have contributed to this outcome include prior variability different from nominal values, low-level sensorimotor learning with the narrow prior, or departures of pursuit from strict adherence to reliability-based weighting. Although modifications of, or alternatives to, the normative Bayesian model will be required, these results, along with previous studies, suggest that Bayesian approaches are a promising framework to understand how pursuit combines immediate sensory motion, past history, and informative perceptual cues to accurately track the target motion that is most likely to occur in the immediate future.NEW & NOTEWORTHY Smooth pursuit eye movements respond on the basis of anticipated, as well as immediate, target motions. Bayesian models using reliability-based weighting of previous (prior) and immediate target motions (likelihood) accounted for many, but not all, aspects of pursuit of clear and noisy random dot kinematograms with different levels of predictability. Bayesian approaches may solve the long-standing problem of how pursuit combines immediate sensory motion and anticipation of future motion to configure an effective response.

Normative tDCS over V5 and FEF reveals practice-induced modulation of extraretinal smooth pursuit mechanisms, but no specific stimulation effect.

The neural networks subserving smooth pursuit eye movements (SPEM) provide an ideal model for investigating the interaction of sensory processing and motor control during ongoing movements. To better understand core plasticity aspects of sensorimotor processing for SPEM, normative sham, anodal or cathodal transcranial direct current stimulation (tDCS) was applied over visual area V5 and frontal eye fields (FEF) in sixty healthy participants. The identical within-subject paradigm was used to assess SPEM modulations by practice. While no specific tDCS effects were revealed, within- and between-session practice effects indicate plasticity of top-down extraretinal mechanisms that mainly affect SPEM in the absence of visual input and during SPEM initiation. To explore the potential of tDCS effects, individual electric field simulations were computed based on calibrated finite element head models and individual functional localization of V5 and FEF location (using functional MRI) and orientation (using combined EEG/MEG) was conducted. Simulations revealed only limited electric field target intensities induced by the applied normative tDCS montages but indicate the potential efficacy of personalized tDCS for the modulation of SPEM. In sum, results indicate the potential susceptibility of extraretinal SPEM control to targeted external neuromodulation (e.g., personalized tDCS) and intrinsic learning protocols.

Tracking a moving visual target in the rhesus monkey: influence of the occurrence frequency of the target path.

Following previous studies documenting the ability to generate anticipatory responses, we tested whether the repeated motion of a visual target along the same path affected its oculomotor tracking. In six rhesus monkeys, we evaluated how the frequency of a target path influenced the onset, accuracy, and velocity of eye movements. Three hundred milliseconds after its extinction, a central target reappeared and immediately moved toward the periphery in four possible (oblique) directions and at a constant speed (20°/s or 40°/s). During each daily session, the frequency of one motion direction was either uncertain (25% of trials) or certain (100% of trials). Our results show no reduction of saccade latency between the two sessions. No express saccades were observed in either session. A slow eye movement started after target onset (presaccadic glissade) and its velocity was larger during the "certain" sessions only with the 40°/s target. No anticipatory eye movement was observed. Longer intersaccadic intervals were found during the "certain" sessions but the postsaccadic pursuit velocity exhibited no change. No correlation was found between the accuracy and precision of saccades (interceptive or catch-up) and the postsaccadic pursuit velocity. Repeatedly tracking a target that moves always along the same path does not favor the generation of anticipatory eye movements, saccadic or slow. Their occurrence is not spontaneous but seems to require preliminary training. Finally, for both sessions, the lack of correlation between the saccade-related and pursuit-related kinematic parameters is consistent with separate control of saccadic and slow eye movements.NEW & NOTEWORTHY Following previous studies documenting anticipatory movements, we investigated how the frequency of occurrence of a target path influenced the generation of tracking eye movements. When present, the effects were small. The limited performance that we found suggests that anticipatory responses require preliminary training, in which case, they should not be considered as a behavioral marker of the primates' ability to extrapolate but the outcome of learning and remembering past experience.

Objects guide human gaze behavior in dynamic real-world scenes.

The complexity of natural scenes makes it challenging to experimentally study the mechanisms behind human gaze behavior when viewing dynamic environments. Historically, eye movements were believed to be driven primarily by space-based attention towards locations with salient features. Increasing evidence suggests, however, that visual attention does not select locations with high saliency but operates on attentional units given by the objects in the scene. We present a new computational framework to investigate the importance of objects for attentional guidance. This framework is designed to simulate realistic scanpaths for dynamic real-world scenes, including saccade timing and smooth pursuit behavior. Individual model components are based on psychophysically uncovered mechanisms of visual attention and saccadic decision-making. All mechanisms are implemented in a modular fashion with a small number of well-interpretable parameters. To systematically analyze the importance of objects in guiding gaze behavior, we implemented five different models within this framework: two purely spatial models, where one is based on low-level saliency and one on high-level saliency, two object-based models, with one incorporating low-level saliency for each object and the other one not using any saliency information, and a mixed model with object-based attention and selection but space-based inhibition of return. We optimized each model's parameters to reproduce the saccade amplitude and fixation duration distributions of human scanpaths using evolutionary algorithms. We compared model performance with respect to spatial and temporal fixation behavior, including the proportion of fixations exploring the background, as well as detecting, inspecting, and returning to objects. A model with object-based attention and inhibition, which uses saliency information to prioritize between objects for saccadic selection, leads to scanpath statistics with the highest similarity to the human data. This demonstrates that scanpath models benefit from object-based attention and selection, suggesting that object-level attentional units play an important role in guiding attentional processing.