Article From Exploration to Fixation: An Integrative View of Yarbus’s Vision Susana Martinez-Conde and Stephen L. Macknik Department of Ophthalmology, State University of New York, Downstate Medical Center, Brooklyn, NY, USA Abstract Alfred Lukyanovich Yarbus (1914–1986) pioneered the study of stabilized retinal images, miniature eye movements, and the cognitive influences that act on visual scanning. Yarbus’s studies of these different topics have remained fundamentally disconnected and independent of each other, however. In this review, we propose that Yarbus’s various research lines are instead deeply and intrinsically interconnected, as are the small eye movements produced during visual fixation and the large-scale scanning patterns associated with visual exploration of objects and scenes. Such apparently disparate viewing behaviors may represent the extremes of a single continuum of oculomotor performance that operates across spatial scales when we search the visual world. Keywords Yarbus, saccades, microsaccades, fixation, visual search, fixational eye movements Introduction Contemporary research on eye movements and vision owes much of its foundation to the work of Alfred Lukyanovich Yarbus (1914–1986; Wade, 2015). Although the details of Yarbus’s life have remained largely obscure to English readers until recently (Tatler, Wade, Kwan, Findlay, & Velichkovsky, 2010), his work influenced powerfully the field of eye movement research, especially since the 1967 English translation of his book, Eye Movements and Vision, originally published in Russian in 1965 (Yarbus, 1967). Yarbus’s work on stabilized retinal images (Figure 1) and on the cognitive influences on scanning patterns (Figures 2 and 3) have each had a very strong impact on current oculomotor research (Tatler et al., 2010). These two research lines represent viewing conditions that are polar opposites to each other in a number of ways: (1) Vision in the absence of eye movements versus vision with unrestricted eye movements. Corresponding author: Susana Martinez-Conde, Department of Ophthalmology, State University of New York, Downstate Medical Center, 450 Clarkson Avenue, Brooklyn, NY 11203, USA. Email: [email protected]Perception 2015, Vol. 44(8–9) 884–899 ! The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0301006615594963 pec.sagepub.com
Transcript
Article
From Exploration to Fixation:An Integrative View ofYarbus’s Vision
Susana Martinez-Conde and Stephen L. MacknikDepartment of Ophthalmology, State University of New York,
Downstate Medical Center, Brooklyn, NY, USA
Abstract
Alfred Lukyanovich Yarbus (1914–1986) pioneered the study of stabilized retinal images, miniature
eye movements, and the cognitive influences that act on visual scanning. Yarbus’s studies of these
different topics have remained fundamentally disconnected and independent of each other,
however. In this review, we propose that Yarbus’s various research lines are instead deeply and
intrinsically interconnected, as are the small eye movements produced during visual fixation and
the large-scale scanning patterns associated with visual exploration of objects and scenes. Such
apparently disparate viewing behaviors may represent the extremes of a single continuum of
oculomotor performance that operates across spatial scales when we search the visual world.
Contemporary research on eye movements and vision owes much of its foundation to thework of Alfred Lukyanovich Yarbus (1914–1986; Wade, 2015). Although the details ofYarbus’s life have remained largely obscure to English readers until recently (Tatler,Wade, Kwan, Findlay, & Velichkovsky, 2010), his work influenced powerfully the field ofeye movement research, especially since the 1967 English translation of his book, EyeMovements and Vision, originally published in Russian in 1965 (Yarbus, 1967).
Yarbus’s work on stabilized retinal images (Figure 1) and on the cognitive influences onscanning patterns (Figures 2 and 3) have each had a very strong impact on currentoculomotor research (Tatler et al., 2010). These two research lines represent viewingconditions that are polar opposites to each other in a number of ways:
(1) Vision in the absence of eye movements versus vision with unrestricted eye movements.
Corresponding author:
Susana Martinez-Conde, Department of Ophthalmology, State University of New York, Downstate Medical Center,
(2) Vision in highly contrived laboratory conditions versus an approximation to everydayvision—through the use of natural images and ecologically relevant viewing tasks.
(3) Microscopic eye movements versus large saccades.
Yet, such apparently disparate matters are deeply and intrinsically connected. Indeed, theymay represent the two extremes of a continuum of oculomotor behavior that encompassesboth vast and minute scanning of the visual world.
The Fixation Misnomer
Classic studies by Yarbus and other eye movement pioneers were instrumental to thecontemporary view that there is no such thing as strict visual fixation. Even when we gazeintently upon a defined target, our eyes never stop moving. The earliest references to thisconstant eye motion date back to Jurin, who in 1738 wrote about the ‘‘trembling of the eye’’(Jurin, 1738), and Helmholtz, who some decades later concurred that it is nearly impossibleto maintain precise fixation (Helmholtz, 1985). Studies conducted from the 1800s through the1950s further established the existence of three main types of small eye movements during therelatively stable ‘‘fixation’’ periods that occur between saccades: These tiny motions areknown today as microsaccades (or fixational saccades), drifts, and tremor. Microsaccades,the largest and fastest of the three fixational eye movements, are involuntary, jerk-likemotions with comparable dynamics to those of classical saccadic eye movements. Drift, aslow, curvy motion of the eye, occurs between microsaccades. Tremor (or ocularmicrotremor), the smallest of the three, takes the form of a very rapid oscillation thatoccurs concurrently with drift (Figure 4).
Starting in the 1950s, several investigators, including Yarbus, characterized visualperception in the absence of all eye movements, including fixational eye movements
Figure 1. Schematic of the suction cup technique used by Yarbus. The target image is directly attached to
the eyeball by means of a tightly-fitting contact lens assembly. The target is viewed through a powerful lens.
The assembly is firmly attached to the eye by a suction device. Source: Martinez-Conde et al. (2004).
Martinez-Conde and Macknik 885
(Barlow, 1963; Ditchburn & Ginsborg, 1952; Riggs & Ratliff, 1952; Yarbus, 1957). Much ofthis research relied on retinal stabilization techniques, where researchers shifted visual stimuliin such a way as to effectively cancel all eye movements. One way to accomplish this was toattach the visual target directly to the eyeball; thus, every time there was a gaze displacement,the stimulus moved with the eye, therefore negating the effect of eye movements and keepingthe retinal images stable. Yarbus’s retinal stabilization studies relied on a novelimplementation of the suction cup technique (Wade, 2015); this consisted of a suctiondevice that served to attach a tightly fitting contact lens assembly to the subject’s eyeball.The contact lens assembly displayed a target image, which the subject viewed through apowerful lens (Figure 1).
Early retinal stabilization studies by Yarbus and others showed that eye movementcancellation led to visual fading: Some amount of time (usually the scale of seconds,but sometimes longer) after retinal stabilization, visual perception faded ontoa homogeneous field. When the eyes were released from stabilization, or if the stabilizedimage changed, perception quickly reappeared (Ditchburn, Fender, & Mayne, 1959;Drysdale, 1975; Gerrits & Vendrik, 1970; Krauskopf, 1957; Sharpe, 1972). Contact lensslippage and other technical difficulties may have produced imperfect retinal stabilization,
Figure 2. ‘‘The unexpected visitor.’’ The same subject inspected the top left image seven times, following
different instructions in each viewing: (a) free examination of the picture; (b) estimate the material
circumstances of the family in the picture; (c) give the ages of the people; (d) surmise what the family had been
doing before the arrival of the ‘‘unexpected visitor’’; (e) remember the clothes worn by the people; (f)
remember the position of the people and objects in the room; (g) estimate how long the ‘‘unexpected visitor’’
had been away from the family. Source: Yarbus (1967).
886 Perception 44(8–9)
Figure 3. Eye-position traces (45 s each) during various viewing tasks. Different combinations of visual
stimuli and viewing task result in different exploration strategies. (a) Visual exploration of a blank scene is
sluggish and uneven. The subject’s gaze tends to remain near the center of the screen. (b) Visual exploration
of a natural scene. Eye fixations concentrate on salient parts of the image (such as faces vs. non-faces, and
foreground vs. background). (c) Picture puzzle visual search. Large horizontal saccades are predominant,
linking equivalent points in the two near-identical images. (d) Where’s Waldo search task. There are higher
concentrations of fixations over the two identified targets (‘‘Waldo’’ and ‘‘Wenda’’ characters). Source:
Figure 4. Fixational eye movement trajectories across the photoreceptor mosaic. Microsaccades are fast
jerk-like movements (straight lines). High-frequency tremor is superimposed on slow drifts (curved lines).
Source: Pritchard (1961).
Martinez-Conde and Macknik 887
however, and thus contributed to the slow onset of visual fading (and occasionalreappearance of the faded images) in early retinal stabilization research (Barlow, 1952,1963). More recently, Coppola and Purves (1996) found that the entoptic images formedby the retinal vasculature (which are stable with respect to the eye as they are directlyattached to the retina) can disappear in less than 80ms.
Such retinal stabilization studies by Yarbus and others established that fixational eyemovements are critical to maintain unchanging objects visible during fixation (Martinez-Conde, Macknik, & Hubel, 2004).
We note that both fixational and nonfixational eye movements play many other roles inhuman vision besides preventing and counteracting perceptual fading: Some of thesefunctions include the foveation and refoveation of targets of interest, visual exploration,and fixation correction, among others (Martinez-Conde, Otero-Millan, & Macknik, 2013;McCamy, Macknik, & Martinez-Conde, in press; McCamy et al., 2012).
The Saccadic Eye Movement Continuum
Saccades are rapid eye movements that change the line of sight between successive points offixation (Figures 2 and 3). They encompass a range of behaviors, including both voluntaryand involuntary shifts of fixation, as well as the quick phases of vestibular and optokineticnystagmus, and the rapid eye movements that occur during sleep (McCamy et al., 2013).
Recent research indicates that saccades of all sizes share a common generator. Both largeand small saccades—including those that interrupt accurate fixation—are part of a saccadiccontinuum that operates across spatial scales (Martinez-Conde et al., 2013; Otero-Millan,Macknik, Langston, & Martinez-Conde, 2013).
The evidence supporting a saccadic continuum includes data from both behavioral andphysiological research. Behavioral studies have shown that microsaccades and saccades sharenumerous physical and functional properties. For instance, both saccades and microsaccadesare typically binocular and conjugate (Ditchburn & Ginsborg, 1953; Krauskopf, Cornsweet,& Riggs, 1960; Lord, 1951) and follow the main sequence (Otero-Millan et al., 2008; Zuber,Stark, & Cook, 1965; Figure 5(a)). The temporal interactions between saccades andmicrosaccades further suggest a common oculomotor generator (i.e., if saccades andmicrosaccades share a common generator, it follows that microsaccade production shouldaffect the timing of saccade production, and vice versa). Saccades that occur shortly aftermicrosaccades have longer latencies than saccades that are not close in time to microsaccades(Rolfs, Laubrock, & Kliegl, 2006, 2008). In addition, there are equivalent time intervalsbetween microsaccades and saccades during exploration and search (Otero-Millan et al.,2008; Figure 5(b)), indicating that saccades and microsaccades share timing constraints(i.e., there are comparable refractory periods from microsaccades to microsaccades,saccades to saccades, microsaccades to saccades, and saccades to microsaccades). Theseobservations are consistent with the hypothesis of a shared saccade and microsaccadegeneration mechanism (Otero-Millan et al., 2008).
The continuum from microsaccades to saccades may extend to ‘‘saccadic intrusions’’(Gowen, Abadi, Poliakoff, Hansen, & Miall, 2007; Martinez-Conde et al., 2013; Otero-Millan et al., 2011)—involuntary saccades that intrude on precise fixation and that areprevalent in various neurological disorders. Recent research has proposed that althoughmicrosaccades and square-wave jerks (SWJs)—the most common type of saccadicintrusion, consisting on a horizontal saccade away from the target and followed by areturn saccade shortly thereafter—are traditionally considered as different types of eyemovements, they are in fact fundamentally the same eye movement with different names
888 Perception 44(8–9)
(Gowen et al., 2007; Otero-Millan, Schneider, Leigh, Macknik, & Martinez-Conde, 2013;Otero-Millan et al., 2011). Otero-Millan et al. (2011) proposed that the apparently dissimilarfeatures of microsaccades and SWJs may result from two complementary mechanisms: one toproduce microsaccades and another to correct fixation errors (via a return microsaccade)when a given microsaccade is too large. This would produce the square-wave coupling seenfor pairs of microsaccades in normal vision (Otero-Millan et al., 2011), and also the clinicalSWJs observed in neurological patients suffering from Parkinson’s disease and othermovement disorders (Otero-Millan, Schneider, et al., 2013; Otero-Millan et al., 2011).
Neural recordings from primate oculomotor structures have provided further evidence fora shared generator for microsaccades and saccades. Pioneering studies showed that burstneurons in the pontomedullary reticular formation (downstream from the superior colliculus[SC]) and putative motor neurons in the nearby abducens nucleus are active during bothmicrosaccades and saccades (Van Gisbergen & Robinson, 1977; Van Gisbergen, Robinson, &Gielen, 1981). Similarly, neurons in the SC rostral pole (which represents foveal goallocations) are as active for small saccades as neurons in the SC caudal region (whichrepresents peripheral goal locations) are for large saccades (Munoz & Wurtz, 1993).Recent research has also found equivalent neural activity around saccades andmicrosaccades in numerous brain structures. The activity of omnipause neurons in thepontine raphe decreases during microsaccades (Brien, Corneil, Fecteau, Andrew, &Douglas, 2009; Van Horn & Cullen, 2012), and there is a continuous representation ofsaccade directions and amplitudes through the SC, down to the smallest microsaccades(Hafed, Goffart, & Krauzlis, 2009), with microsaccade representation in the SC rostralpole, in agreement with previous observations (Munoz & Wurtz, 1993). Neural activityduring microsaccades sometimes extends to small-amplitude voluntary saccades, consistentwith behavioral evidence of a saccadic continuum, and rostral SC inactivation results indecreased microsaccade rates (Hafed et al., 2009). Unilateral inactivation of the fastigialoculomotor region in the cerebellum likewise affects the metrics of visually guided saccades
Figure 5. Equivalent functional dynamics of microsaccades and saccades. (a): Microsaccades and saccades
during free viewing (blue) follow the same main sequence as during attempted fixation (red). Some of the red
dots are obscuring the blue dots underneath. (b): Intersaccadic intervals follow similar distributions for all
saccade–microsaccade combinations (N¼ 8 subjects). The only variation between distributions occurs for
intersaccadic intervals larger than 200 ms. Source: Otero-Millan et al. (2008).
Microsaccades During Visual Exploration and Search
Microsaccades were originally observed, and described, during the attempt to maintainprolonged fixation. Around the turn of the millennium, researchers wondered whetherviewing conditions other than straight fixation on a target (such as the visual explorationand visual search of natural images) might also produce microsaccadic eye movements. Anearly attempt to investigate microsaccade production in freely moving human subjects(Malinov, Epelboim, Herst, & Steinman, 2000) concluded that only 2 out of >3,000 totalsaccades recorded could be classified as ‘‘microsaccades.’’ However, this study definedmicrosaccades in very stringent terms (i.e., saccades with magnitudes <12 arcmin) thatwere not in line with contemporary reports of microsaccade dynamics. (The vast majorityof search coil and high-resolution video tracking recordings of human microsaccadesconducted in the last two decades have reported considerably larger microsaccademagnitudes during attempted fixation, up to 1� to 2�, depending on the precise task andother experimental conditions.) Some years later, Otero-Millan et al. (2008) revisited thequestion of microsaccade production during free-viewing. To do this, they considered allsaccades below 1� as microsaccades, irrespective of whether they were produced duringprolonged fixation, or during the brief fixation periods in between large saccades duringnon-fixation tasks. Their results showed considerable microsaccade production acrossmultiple viewing tasks, including visual exploration, visual search, and prolonged fixation,therefore indicating a role of microsaccades in natural vision and free viewing, for the first time(Figure 6; See also, Martinez-Conde, 2006 and Figure 7). Otero-Millan et al. (2008) also foundthat saccades andmicrosaccades had comparable spatiotemporal characteristics, including thepresence of equivalent refractory periods between all pair-wise combinations of saccades andmicrosaccades, consistent with the concept of a microsaccade–saccade continuum (Figure5(b)). More recently, McCamy et al. (2014) found that observers generated moremicrosaccades on more informative than less informative regions of natural images.Increased microsaccade production was not fully explained by increased fixation duration,however, suggesting that the visual system uses microsaccades to heighten informationacquisition from informative regions. Other studies have since concurred that microsaccadesoccur during free viewing, including in simulation scenarios representative of ecologicalconditions (Benedetto, Pedrotti, & Bridgeman, 2011; Di Stasi et al., 2013, 2015).
A Fixation–Exploration Continuum
The hypothetical separation between exploratory gaze shifts and attempted fixation dates backto the discovery of fixational eye movements (Ditchburn & Ginsborg, 1953; Dodge, 1907;Ratliff & Riggs, 1950) and remained a fixture in visual, cognitive, and oculomotor researchfor many decades (Otero-Millan et al., 2008). Classic vision studies, including those by Yarbus,distinguished between visual exploration—characterized by the alternation of saccades andbrief fixation periods—and attempted visual fixation—where observers maintained relativegaze stability despite the continuous occurrence of fixational eye movements (Barlow, 1952;Ditchburn & Ginsborg, 1953; Dodge, 1907; Ratliff & Riggs, 1950; Rolfs, 2009; Yarbus, 1967).Mounting evidence in support of a common generator for exploratory saccades and fixationalmicrosaccades nevertheless brought into question a dichotomy between exploration andfixation. Instead, it is possible that saccades and microsaccades form an oculomotor
890 Perception 44(8–9)
continuum along the spectrum of spatial scales, with classical exploratory saccades at one endand classical fixational microsaccades at the other end.
To be certain of such a conclusion, one must first reconcile the known discrepanciesbetween the dynamics of saccades during exploration and those of microsaccades duringfixation. One difference is that, during exploration, saccades occur at a rate of about threeper second, whereas during fixation, microsaccades occur about once per second (Martinez-Conde, Macknik, Troncoso, & Hubel, 2009; Otero-Millan et al., 2008; Rolfs, 2009). Otero-Millan, Macknik, et al. (2013) proposed that this paradox would be resolved if explorationand fixation are not opposing behaviors, but instead the two extremes of a functionalcontinuum in which saccades scan visual scenes of any sizes, including the smallest. Toaddress this possibility, they tracked the eye movements of human participants while theyfixated a small dot or freely explored natural images and blank scenes with sizes ranging fromthe massive to the minute (160� to 0.5� of visual angle).
They found that saccadic rates not only decreased as the size of the field of explorationdecreased but moreover fell on a continuum (Figure 8). Other saccadic properties, includingdirection, magnitude, peak velocity, and intersaccadic interval, also varied as function ofimage size, forming a continuum with the parameters of microsaccades during fixation. Theseresults suggested that the human oculomotor system engages in continuous exploration whileobserving objects of all dimensions, with the size of the area of exploration determiningsaccadic parameters such as rate and magnitude. In other words, when people observesmall things (i.e., minute scenes, object features, fixation targets), their oculomotor systemsscan them as they do the entire visual scene during a classic exploration scenario, but withsmall and infrequent saccades, rather than the larger and more frequent saccades generatedduring classic exploration conditions.
Figure 6. Microsaccades and saccades during free viewing. Left: A subject’s eye positions during free visual
exploration. Right: 10-s period of the same subject’s eye position from the left panel, in more detail. Faces
were a primary focus of fixation (as previously found by Yarbus, 1967) and also concentrated abundant
microsaccades. Source: Martinez-Conde et al. (2013).
Martinez-Conde and Macknik 891
Discrete Sampling at Multiple Spatial Scales
The spatiotemporal dynamics of saccades and microsaccades might reflect a samplingstrategy by which the brain discretely acquires visual information at multiple spatial scales.Gilchrist et al. observed that a patient (AI) who was unable to make eye movements (except
Figure 7. ‘‘Girl from the Volga.’’ Top right: The scan patterns represent one subject’s free examination of
the photograph during 3 min. Source: (Yarbus, 1967). Bottom left: Replication of Yarbus’s study: The scan
patterns represent one subject’s free examination of the photograph during 2 min. Bottom right: The
microsaccades produced in the brief fixation periods during the free scanning follow the main sequence (i.e.,
for small-amplitude drifts) produced head saccades instead (Figure 9). Such head saccadeswere comparable to standard eye saccades in many of their characteristics, including theiramplitude, duration of intersaccadic intervals, length of intervening fixations, and range ofvisual scanning during picture exploration (the peak velocity of head saccades was slowerthan that of regular eye saccades, however). These head saccades enabled the patient to readat normal speed and even carry out complex visuomotor tasks, such as making a cup of tea,without problems. The authors concluded that ‘‘saccadic movements, of the head or the eye,form the optimal sampling method for the brain’’ (compared with smooth scanning of thevisual scene; Gilchrist et al., 1997; Gilchrist, Brown, Findlay, & Clarke, 1998; Land,Furneaux, & Gilchrist, 2002).
Physiological studies comparing the neuronal responses triggered by saccades/microsaccades to the responses triggered by instantaneous events, such as blinks andflashes, further support the idea that saccades and microsaccades sample visualinformation discretely (Martinez-Conde et al., 2009). Gawne and Martin (2002) foundmost neurons in V1, V2, V3V/VP, and V4V to have similar responses to the onsets andterminations of visual stimuli elicited by flashes, blinks, and saccades. This findingsuggested that the neural circuits underlying visual perception respond to various kinds oftransient events in similar fashion (Gawne & Martin, 2002). Martinez-Conde et al. (2002)found that neural responses to flashes in the LGN and V1 were stronger than—but in thesame order of magnitude as—responses to microsaccades, perhaps because of the greaterabruptness of flashes with respect to microsaccades. Transient responses evoked bymicrosaccades in primate visual neurons usually take the form of bursts of spikes
Figure 8. A saccadic continuum from exploration to fixation. Top: Average saccade rates decrease
parametrically with scene size. Error bars represent SEM across subjects (N¼ 10). Bottom: Image examples,
proportionally scaled down from the sizes presented during the experiment. Source: Otero-Millan et al.
(2013).
Martinez-Conde and Macknik 893
(Martinez-Conde, Macknik, & Hubel, 2000; Martinez-Conde et al., 2002, 2004), which mayor may not be followed by sustained firing during intersaccadic periods. Both bursty firingand microsaccade production have been related to visibility during fixation (McCamy, Otero-Millan, et al., 2014; McCamy et al., 2012; Martinez-Conde, 2006; Martinez-Conde et al.,2000, 2002; Martinez-Conde, Macknik, Troncoso, & Dyar, 2006; Troncoso, Macknik, &Martinez-Conde, 2008). In addition, the suppression of transient bursts is linked toperceptual suppression during blinks (Gawne & Martin, 2000) and to decreased targetvisibility during visual masking (Macknik & Livingstone, 1998; Macknik & Martinez-Conde, 2004; Macknik, Martinez-Conde, & Haglund, 2000). Other studies have found thatV1 neurons produce stronger responses to transient than to drifting stimuli and suggestedthat neural transients might underlie the behavior of cortical neurons as coincidence detectors(Shelley, McLaughlin, Shapley, & Wielaard, 2002; Williams & Shapley, 2007). Slow gradualchanges (which presumably result in sustained, rather than transient, firing) are also harder todetect than abrupt changes (Simons, Franconeri, & Reimer, 2000).
The finding that microsaccades help to reverse and prevent perceptual fading provides apsychophysical counterpart to the discovery that microsaccades generate neural transients invisual neurons during fixation. Martinez-Conde et al. (2006) asked subjects to report thevisibility of visual targets that faded and intensified perceptually during fixation (i.e., Troxler
Figure 9. Head movements from patient AI while viewing the ‘‘Girl from the Volga’’ picture during 20 s. The
head position was sampled at 100 Hz. Source: Gilchrist, Brown, & Findlay (1997).
894 Perception 44(8–9)
fading; Figure 10; Martinez-Conde et al., 2006). They found that microsaccade onsets led tothe visual restoration of faded targets, establishing a link between microsaccade productionand visibility. Subsequent research showed that microsaccades are the most important eyemovement contributor to restoring faded vision during fixation, for both foveal andperipheral targets (McCamy et al., 2012) and that both microsaccades and drift worktogether to prevent fading from happening in the first place (McCamy, Macknik, &Martinez-Conde, 2014).
Otero-Millan, Macknik, & Martinez-Conde (2012) also found a strong connectionbetween microsaccade and blink production and the perception of the Rotating SnakesIllusion (Kitaoka, 2005) during fixation. This finding suggested that transient ocularevents, such as microsaccades, saccades, and blinks, can trigger the perception of illusoryrotation in certain static repetitive patterns, and is consistent with the generation of neuraltransients in visual neurons by (micro)saccades and blinks.
Discrete sampling may speed up neural processing by chunking information and thusprovide an evolutionary advantage (Uchida, Kepecs, & Mainen, 2006), not only in visionbut in other sensory systems as well. Sniffs in rodents discretely sample olfactory informationevery 200 to 300ms and thus are similar in their temporal dynamics to saccades andmicrosaccades in primates (Otero-Millan et al., 2008; Uchida et al., 2006). Discretesampling may also apply to tactile information, for instance, when subjects identify anunseen object with their fingertips or when blind people read Braille. Observers candetermine the gist of a briefly flashed scene in a time as short as 150ms, but not shorter(Thorpe, Fize, & Marlot, 1996). The fact that this interval cannot be reduced, even withextensive training, suggests a lower limit in the number of neural stages and speed involved invisual information processing (Fabre-Thorpe, Delorme, Marlot, & Thorpe, 2001). Thus, eventhough the human oculomotor system only generates saccades/microsaccades every couple of100ms, there is reason to think that faster rates of saccade/microsaccade production mightnot improve vision significantly.
Figure 10. Troxler fading demonstration. Fixate your gaze on the red dot while attending to the blue ring
around it. The ring will fade in a few seconds. Microsaccades (and also large saccades and blinks) bring the
faded ring back to visibility. Source: Martinez-Conde et al. (2006).
Martinez-Conde and Macknik 895
Conclusions
Yarbus pioneered the study of perception during retinal stabilization and the investigation ofthe cognitive influences that act on scanning patterns during visual exploration. These tworesearch lines are far from disjointed, and recent work indicates a deeper connection than wasknown in Yarbus’s time. New behavioral and neurophysiological data support the hypothesisof a common generator for exploratory saccades and fixational microsaccades and suggest acontinuum of oculomotor behavior that encompasses both vast and minute scanning of thevisual world.
Acknowledgements
We are grateful to the organizers of the Yarbus-100 conference for inviting S. M. C. to present a
keynote lecture that served as the basis for this article. We also thank Max Dorfman for
administrative assistance and Jordi Chanovas for help with the figures.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or
publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or
publication of this article: This work was supported by a challenge grant from Research to Prevent
Blindness Inc. to the Department of Ophthalmology at SUNY Downstate and by the Empire
Innovation Program (awards to S. M. C. and S. L. M.).
References
Barlow, H. B. (1952). Eye movements during fixation. The Journal of Physiology, 116, 290–306.
Barlow, H. B. (1963). Slippage of contact lenses and other artifacts in relation to fading andregeneration of supposedly stable retinal images. The Quarterly Journal of ExperimentalPsychology, 15, 36–51.
Benedetto, S., Pedrotti, M., & Bridgeman, B. (2011). Microsaccades and exploratory saccdes in a
naturalistic environment. Journal of Eye Movement Research, 4, 1–10.Brien, D. C., Corneil, B. D., Fecteau, J. H., Bell, A. H., & Munoz, P. M. (2009). The behavioural and
neurophysiological modulation of microsaccades in monkeys. Journal of Eye Movement Research, 3,
1–12.Coppola, D., & Purves, D. (1996). The extraordinarily rapid disappearance of entoptic images.
Proceedings of the National Academy of Sciences of the United States of America, 93, 8001–8004.
Di Stasi, L. L., McCamy, M. B., Catena, A., Macknik, S. L., Canas, J. J., & Martinez-Conde, S. (2013).Microsaccade and drift dynamics reflect mental fatigue. European Journal of Neuroscience, 38,2389–2398.
Di Stasi, L. L., McCamy, M. B., Pannasch, S., Renner, R., Catena, A., Canas, J. J., . . .Martinez-Conde, S. (2015). Effects of driving time on microsaccadic dynamics. Experimental BrainResearch, 233, 599–605.
Ditchburn, R. W., Fender, D. H., & Mayne, S. (1959). Vision with controlled movements of the retinal
image. The Journal of Physiology, 145, 98–107.Ditchburn, R. W., & Ginsborg, B. L. (1952). Vision with a stabilized retinal image. Nature, 170, 36–37.
896 Perception 44(8–9)
Ditchburn, R. W., & Ginsborg, B. L. (1953). Involuntary eye movements during fixation. The Journal of
Physiology, 119, 1–17.Dodge, R. (1907). An experimental study of visual fixation. Psychological Monographs: General and
Applied, 8, 1–95.
Drysdale, A. E. (1975). The visibility of retinal blood vessels. Vision Research, 15, 813–818.Fabre-Thorpe, M., Delorme, A., Marlot, C., & Thorpe, S. (2001). A limit to the speed of processing in
ultra-rapid visual categorization of novel natural scenes. Journal of Cognitive Neuroscience, 13,171–180.
Gawne, T. J., & Martin, J. M. (2000). Activity of primate V1 cortical neurons during blinks. Journal ofNeurophysiology, 84, 2691–2694.
Gawne, T. J., & Martin, J. M. (2002). Responses of primate visual cortical neurons to stimuli presented
by flash, saccade, blink, and external darkening. Journal of Neurophysiology, 88, 2178–2186.Gerrits, H. J., & Vendrik, A. J. (1970). Artificial movements of a stabilized image. Vision Research, 10,
1443–1456.
Gilchrist, I. D., Brown, V., & Findlay, J. M. (1997). Saccades without eye movements. Nature, 390,130–131.
Gilchrist, I. D., Brown, V., Findlay, J. M., & Clarke, M. P. (1998). Using the eye-movement system to
control the head. Proceedings of the Royal Society of London, Series B: Biological Sciences, 265,1831–1836.
Goffart, L., Chen, L. L., & Sparks, D. L. (2004). Deficits in saccades and fixation during muscimolinactivation of the caudal fastigial nucleus in the rhesus monkey. Journal of Neurophysiology, 92,
3351–3367.Gowen, E., Abadi, R. V., Poliakoff, E., Hansen, P. C., & Miall, R. C. (2007). Modulation of saccadic
intrusions by exogenous and endogenous attention. Brain Research, 1141, 154–167.
Guerrasio, L., Quinet, J., Buttner, U., & Goffart, L. (2010). Fastigial oculomotor region and the controlof foveation during fixation. Journal of Neurophysiology, 103, 1988–2001.
Hafed, Z. M., Goffart, L., & Krauzlis, R. (2009). A neural mechanism for microsaccade generation in
the primate superior colliculus. Science, 323, 940–943.Helmholtz, H. (1985). Helmholtz’s treatise on physiological optics (J. P. C. Southall, Trans. Translated
from 3rd German ed. Vol. 2). Birmingham, Alabama: Gryphon Editions, Ltd.
Jurin, J. (1738). An essay on distinct and indistinct vision. In R. Smith (Ed.), A compleat system ofopticks in four books (pp. 115–171). Cambridge: Published by the author.
Kitaoka, A. (2002). Trick eyes graphics: Kanzen, Tokyo.Krauskopf, J. (1957). Effect of retinal image motion on contrast thresholds for maintained vision.
Journal of the Optical Society of America, 47, 740–744.Krauskopf, J., Cornsweet, T. N., & Riggs, L. A. (1960). Analysis of eye movements during monocular
and binocular fixation. Journal of the Optical Society of America, 50, 572–578.
Land, M. F., Furneaux, S. M., & Gilchrist, I. D. (2002). The organization of visually mediated actionsin a subject without eye movements. Neurocase, 8, 80–87.
Lord, M. P. (1951). Measurement of binocular eye movements of subjects in the sitting position. British
Journal of Ophthalmology, 35, 21–30.Macknik, S. L., & Livingstone, M. S. (1998). Neuronal correlates of visibility and invisibility in the
primate visual system. Nature Neuroscience, 1, 144–149.Macknik, S. L., & Martinez-Conde, S. (2004). The spatial and temporal effects of lateral inhibitory
networks and their relevance to the visibility of spatiotemporal edges. Neurocomputing, 58–60,775–782.
Macknik, S. L., Martinez-Conde, S., & Haglund, M. M. (2000). The role of spatiotemporal edges in
visibility and visual masking. Proceedings of the National Academy of Sciences of the United States ofAmerica, 97, 7556–7560.
Malinov, I. V., Epelboim, J., Herst, A. N., & Steinman, R. M. (2000). Characteristics of saccades and
vergence in two types of sequential looking tasks. Vision Research, 40, 2083–2090.Martinez-Conde, S. (2006). Fixational eye movements in normal and pathological vision. Progress in
Brain Research, 154, 151–176.
Martinez-Conde and Macknik 897
Martinez-Conde, S., Macknik, S. L., & Hubel, D. H. (2000). Microsaccadic eye movements and firing
of single cells in the striate cortex of macaque monkeys. Nature Neuroscience, 3, 251–258.Martinez-Conde, S., Macknik, S. L., & Hubel, D. H. (2002). The function of bursts of spikes during
visual fixation in the awake primate lateral geniculate nucleus and primary visual cortex. Proceedings
of the National Academy of Sciences of the United States of America, 99, 13920–13925.Martinez-Conde, S., Macknik, S. L., & Hubel, D. H. (2004). The role of fixational eye movements in
visual perception. Nature Reviews Neuroscience, 5, 229–240.Martinez-Conde, S., Macknik, S. L., Troncoso, X., & Dyar, T. A. (2006). Microsaccades counteract
visual fading during fixation. Neuron, 49, 297–305.Martinez-Conde, S., Macknik, S. L., Troncoso, X. G., & Hubel, D. H. (2009). Microsaccades: A
neurophysiological analysis. Trends in Neurosciences, 32, 463–475.
Martinez-Conde, S., Otero-Millan, J., & Macknik, S. L. (2013). The impact of microsaccades on vision:Towards a unified theory of saccadic function. Nature Reviews Neuroscience, 14, 83–96.
McCamy, M. B., Macknik, S. L., & Martinez-Conde, S. (2013). Natural eye movements and vision. In
J. S. Werner & L. M. Chalupa (Eds.), The new visual neurosciences (pp. 849–863). Cambridge, MA:The MIT Press.
McCamy, M. B., Macknik, S. L., & Martinez-Conde, S. (2014). Different fixational eye movements
mediate the prevention and the reversal of visual fading. The Journal of Physiology, 592,4381–4394.
McCamy, M. B., Otero-Millan, J., Macknik, S. L., Yang, Y., Troncoso, X. G., Baer, S.M., . . .Martinez-Conde, S. (2012). Microsaccadic efficacy and contribution to foveal and
peripheral vision. Journal of Neuroscience, 32, 9194–9204.McCamy, M. B., Otero-Millan, J., Di Stasi, L. L., Macknik, S. L., & Martinez-Conde, S. (2014). Highly
informative natural scene regions increase microsaccade production during visual scanning. The
Journal of Neuroscience, 34, 2956–2966.Munoz, D. P., & Wurtz, R. H. (1993). Fixation cells in monkey superior colliculus. I. Characteristics of
cell discharge. Journal of Neurophysiology, 70, 559–575.
Otero-Millan, J., Macknik, S. L., Langston, R. E., & Martinez-Conde, S. (2013). An oculomotorcontinuum from exploration to fixation. Proceedings of the National Academy of Sciences, 110,6175–6180.
Otero-Millan, J., Macknik, S. L., & Martinez-Conde, S. (2012). Microsaccades and blinks triggerillusory rotation in the ‘‘rotating snakes’’ illusion. Journal of Neuroscience, 32, 6043–6051.
Otero-Millan, J., Schneider, R., Leigh, R. J., Macknik, S. L., & Martinez-Conde, S. (2013). Saccadesduring attempted fixation in parkinsonian disorders and recessive ataxia: From microsaccades to
square-wave jerks. PLoS One, 8, 1–9.Otero-Millan, J., Serra, A., Leigh, R. J., Troncoso, X. G., Macknik, S. L., & Martinez-Conde, S.
(2011). Distinctive features of saccadic intrusions and microsaccades in progressive supranuclear
palsy. Journal of Neuroscience, 31, 4379–4387.Otero-Millan, J., Troncoso, X., Macknik, S. L., Serrano-Pedraza, I., & Martinez-Conde, S. (2008).
Saccades and microsaccades during visual fixation, exploration and search: Foundations for a
common saccadic generator. Journal of Vision, 8, 21.1–21.18.Pritchard, R. M. (1961). Stabilized images on the retina. Scientific American, 204, 72–78.Ratliff, F., & Riggs, L. A. (1950). Involuntary motions of the eye during monocular fixation. Journal of
Experimental Psychology, 40, 687–701.
Riggs, L. A., & Ratliff, F. (1952). The effects of counteracting the normal movements of the eye. Journalof the Optical Society of America, 42, 872–873.
Robinson, F. R., Straube, A., & Fuchs, A. F. (1993). Role of the caudal fastigial nucleus in saccade
generation. II. Effects of muscimol inactivation. Journal of Neurophysiology, 70, 1741–1758.Rolfs, M. (2009). Microsaccades: Small steps on a long way. Vision Research, 49, 2415–2441.Rolfs, M., Laubrock, J., & Kliegl, R. (2006). Shortening and prolongation of saccade latencies
following microsaccades. Experimental Brain Research, 169, 369–376.Rolfs, M., Laubrock, J., & Kliegl, R. (2008). Microsaccade-induced prolongation of saccadic latencies
depends on microsaccade amplitude. Journal of Eye Movement Research, 1, 1, 1–8.
898 Perception 44(8–9)
Sharpe, C. R. (1972). The visibility and fading of thin lines visualized by their controlled movement
across the retina. Journal of Physiology (London), 222, 113–134.Shelley, M., McLaughlin, D., Shapley, R., & Wielaard, J. (2002). States of high conductance in a large-
scale model of the visual cortex. Journal of Computational Neuroscience, 13, 93–109.
Simons, D. J., Franconeri, S. L., & Reimer, R. L. (2000). Change blindness in the absence of a visualdisruption. Perception, 29, 1143–1154.
Tatler, B. W., Wade, N. J., Kwan, H., Findlay, J. M., & Velichkovsky, B. M. (2010). Yarbus, eyemovements, and vision. i-Perception, 1, 7.
Thorpe, S., Fize, D., & Marlot, C. (1996). Speed of processing in the human visual system. Nature, 381,520–522.
Troncoso, X., Macknik, S. L., & Martinez-Conde, S. (2008). Microsaccades counteract perceptual
filling-in. Journal of Vision, 8, 15.1–15.9.Uchida, N., Kepecs, A., & Mainen, Z. F. (2006). Seeing at a glance, smelling in a whiff: Rapid forms of
Van Gisbergen, J. A., Robinson, D. A., & Gielen, S. (1981). A quantitative analysis of generation ofsaccadic eye movements by burst neurons. Journal of Neurophysiology, 45, 417–442.
Van Gisbergen, J. A. M., & Robinson, D. A. (1977). Generation of micro and macrosaccades by burst
neurons in the monkey. In R. Baker, & A. Berthoz (Eds.), Control of gaze by brain stem neurons.New York, NY: Elsevier/North-Holland.
Van Horn, M. R., & Cullen, K. E. (2012). Coding of microsaccades in three-dimensional space bypremotor saccadic neurons. Journal of Neuroscience, 32, 1974–1980.
Wade, N. J. (2015). How were eye movements recorded before Yarbus? Perception (This Issue).Williams, P. E., & Shapley, R. M. (2007). A dynamic nonlinearity and spatial phase specificity in
macaque V1 neurons. Journal of Neuroscience, 27, 5706–5718.
Yarbus, A. L. (1957). The perception of an image fixed with respect to the retina. Biophysics, 2,683–690.
Yarbus, A. L. (1967). Eye movements and vision (B. Haigh, Trans.). New York, NY: Plenum Press.
Zuber, B. L., Stark, L., & Cook, G. (1965). Microsaccades and the velocity-amplitude relationship forsaccadic eye movements. Science, 150, 1459–1460.
