The aim of this page is top give a brief description of the different main types of eye movements and their function.
The spatial and temporal sampling ability of the human eye limits the manner in which we extract visual information from events in the world. Because visual acuity decreases rapidly when we move away from the center of our visual field, we possess a repertoire of eye movements that allow us to point our eyes at target locations of interest.
Fixations are the most common feature of looking that eye tracking researchers analyze to make inferences about cognitive processes or states that they are interested in probing. Fixations are those times when our eyes essentially stop scanning about the scene, holding the central foveal vision in place so that the visual system can take in detailed information about what is being looked at. But how do we get fixations? Let’s start with the raw material from which fixations are built: gaze points. Gaze points are the instantaneous spatial locations of the visual axis landing on the stimulus. As such, they have an (x, y) coordinate and a timestamp corresponding to its measurement. Think of gaze points as an output of the eye tracking hardware. If the device is operating at 60 Hz, a gaze point will be reported every 16.7 milliseconds. At 300 Hz, gaze points are spaced a mere 3 milliseconds apart. Thus, the number of gaze points is partially an artifact of the eye tracker and bears little to no direct relation to the things that researchers are interested in. Fixations have two characteristics distinguishing them from gaze points. The first is that since they are made up of multiple gaze points, fixations have duration in addition to a spatial (x, y) location and start and end timestamps. The second is that fixations are not real in the sense of being directly measurable. Fixations are constructions, outputs of a mathematical algorithm that translates the sequence of raw gaze points into an associated sequence of fixations. Paradoxically, fixations are real in the sense that they are meaningful episodes of looking generated by our visual system. These episodes have specific dynamic characteristics that the gaze point-to-fixation conversion algorithm, or fixation filter, is designed to model. So, putting the eye tracker-based raw gaze stream through the fixation filter is an attempt to reconstruct these meaningful eye movements as faithfully as possible. As noted earlier, they have characteristics that can reveal useful information about attention, visibility, mental processing, and understanding. For example, an increase in the time taken to make a first fixation on a target suggests a decrease in the salience or visual attractive power of that feature. An increase in average fixation duration on a target or area could signal greater effort required to make sense of something or could suggest that what is looked at is more engaging. The underlying goal of all eye tracking researchers is to identify which ingredients (eye tracking metrics) to use in building a recipe for the study of the attentional, cognitive states, or processes in which they are interested.
Keep in mind that this recipe could include multiple eye tracking measures and potentially additional triangulating data streams as well. For example, to bolster judgments about cognitive effort, one might simultaneously measure pupil dilation or electrodermal activity (i.e., skin conductance or sweat response). Questionnaire instruments, self-reports, or aided/unaided recall tasks could all serve as useful ingredients in a research recipe that yields strong, defensible conclusions.
Saccades are the type of eye movement used to move the fovea rapidly from one point of interest to another, while a fixation is the period of time where the eye is kept aligned with the target for a certain duration, allowing for the image details to be processed. Our perception is guided by alternating these sequences of fixations and saccades (see figure on the left). Due to the fast movement during a saccade, the image on the retina is of poor quality and information intake thus happens mostly during the fixation period.
Saccade facts:
- can be triggered voluntarily or involuntarily
- both eyes move in the same direction
- the time to “plan” a saccade (latency) is task dependent and varies between 100-1000 ms
- the average duration of a saccade is 20-40 ms
- the duration of a saccade and its amplitude are linearly correlated, i.e. larger jumps produce longer durations
- the end point of a saccade cannot be changed when the eye is moving
- Saccades do not always have simple linear trajectories
Fixation facts:
- a fixation is composed of slower and minute movements (microsaccades, tremor and drift) that help the eye align with the target and avoid perceptual fading (fixational eye movements)
- the duration varies between 50-600 ms (however longer fixations have been reported)
- the minimum duration required for information intake depends on the task and stimulus
Dynamic stimuli and "real world" recordings
When we look at a static object with our heads relatively still, we mainly perform saccades and fixational eye movements. However in more dynamic situations where either we are moving, or the object itself is moving, other eye movements are triggered to keep the fovea aligned with the point of interest. Vergence movements are recruited to help us focus on objects placed at different distances, smooth pursuit is used to keep the fovea aligned with moving objects and the vestibular ocular reflex is used to maintain our fovea pointed at a point of interest when our head and body are moving.
Vergence facts:
- the left and right eye move in opposite directions
- can be classified into two types of movements - far-to-near focus triggers convergent movements and near-to-far focus triggers divergent movements
- are generally slower than saccades
Smooth pursuit facts:
- cannot be triggered voluntarily, in absence of a moving target
- eye velocity is most often less than 30 deg/sec (however some individuals can smooth pursuit at velocities as high as 100 deg/sec)
- when the target moves at a higher speed than 30 deg/sec we start to employ catch up saccades to keep up with the target
Vestibular ocular reflex
- the eyes move in the opposite direction of the head
- normally the speed of the eye equals the speed of the head
Recommended reading
- Rayner, K. 2009. Eye Movements and Attention in Reading, Scene Perception, and Visual Search. Quarterly Journal of Experimental Psychology, 62, 1457-1506. http://dx.doi.org/10.1080/17470210902816461
- Land M, Tatler B. Looking and Acting: Vision and eye movements in natural behaviour. Oxford University Press; 2009. http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780198570943.001.0001/acprof-9780198570943. Accessed March 6, 2018.