topics: |  psychology | gaze | vision |

NJ Wade: Scanning the Seen - origins of eye-movement research

In 1900, Raymond Dodge presented a very early exploration of the possibility
of perception within eye movements. This paper built on earlier observations
made during his time with Benno Erdmann (1851–1921) in which they conducted
experiments concerning vision during reading (Erdmann & Dodge, 1898). They
had been using mirrors to observe subjects’ eye movements whilst reading
text. When looking into the mirrors themselves Erdmann and Dodge both found
it impossible to see their own rapid eye movements.

Until these reports by Erdmann and Dodge, it seems that the general consensus
in the field was that eye movements were themselves an integral part of the
processes of visual perception. It was believed that perception continued
during eye movements and that the continuous movement of gaze over an object
would be sufficient for its perception.  Dodge, however, recognised that this
was not the case. Critically, Dodge appreciated the errors and pitfalls of
self-observation when describing eye movements and perception, in the same
way that Wells had distrusted Porterfield’s recourse to subjective experience
over 100 years earlier.

During fixations the image on the retina is kept largely stationary. When
gaze is relocated, there is a brief period of blindness followed by the
reception of another relatively static view of the new fixation
target. Consequently the input that our eyes supply to the visual system is
in the form of a series of relatively static glimpses of the world separated
by brief periods of blindness. This is very different from the smooth
perception that we experience. Crum Brown appreciated the significance of
saccades and fixations, and described them eloquently: p.52

    We fancy that we can move our eyes uniformly, that by a continuous motion
    like that of a telescope we can move our eyes along the sky-line in the
    landscape or the cornice of a room, but we are wrong in this. However
    determinedly we try to do so, what actually happens is, that our eyes
    move like the seconds hand of a watch, a jerk and a little pause, another
    jerk and so on; only our eyes are not so regular, the jerks are sometimes
    of greater, sometimes of less, angular amount, and the pauses vary in
    duration, although, unless we make an effort, they are always
    short. During the jerks we practically do not see at all, so that we have
    before us not a moving panorama, but a series of fixed pictures of the
    same fixed things, which succeed one another rapidly. (Crum Brown, 1895,
    pp. 4–5)

Dodge and Cline developed a photographic apparatus (1901) to record eye
movements... there followed something of a revolution in eye-movement
research and a proliferation of new experiments in this field (see Taylor,
1937). Other researchers developed similar convenient and effective
eye-trackers and research extended beyond the domain of reading. Researchers
began to consider whether the newly described saccade-and-fixate strategy
applied to tasks other than reading, and it was soon realised that this was
the case.  Dodge: 100 Hz Photochronograph

George Malcolm Stratton (1865–1957) employed a photographic technique to
examine eye movements when viewing patterns. This was an important new
direction for eyemovement research and served to highlight the importance of
saccades outside the context of reading. Like his contemporaries, Stratton
was surprised by the discontinuity of eye movements: “The eye darts from
point to point, interrupting its rapid motion by instants of rest. And the
path by which the eye passes from one to another of these of these resting
places does not seem to depend very nicely upon the exact form of the line
observed” (Stratton, 1902, p. 343).

 
Figure 1. Eye movements made by subjects while examining I. E. Repin’s
painting “An Unexpected Visitor”, with different questions in mind (adapted
from Yarbus, 1967). (a) The picture. (b) “Remember the clothes worn by the
people.” (c) “Remember the positions of the people and objects in the room.”
(d) “Estimate how long the ‘unexpected visitor’ had been away from the
family.” Saccades are the thin lines; fixations are the knot-like
interruptions.

How people look at pictures (Buswell, 1935)

In the monograph, Guy Thomas Buswell (1891–1994), reports eye-movement data
recorded from 200 participants each viewing multiple pictures, such that his
data comprised almost 2000 eye-movement records each containing a large
number of fixations.

 
fixations noted by Buswell for Hokusai's "The wave".  Fixations are numbered
in sequence.

4 Michael F. Land : Fixation Strategies During Active Behaviour

(Yarbus, 1967) demonstrated convincingly that the kinds of eye movements
people make when viewing a scene depend on what information they are trying
to get from it, and not just on the eye-catching power (“intrinsic salience”)
of the objects in that scene (Figure 1). He provided an account of eye
movements in which central control, related to the task in hand, was seen as
being more important than reflex-like responses to stimulus objects.

 
Eye movements made by subjects while examining I. E. Repin’s painting “An
Unexpected Visitor”, with different questions in mind (adapted from Yarbus,
1967). (a) The picture. (b) “Remember the clothes worn by the people.” (c)
“Remember the positions of the people and objects in the room.” (d) “Estimate
how long the ‘unexpected visitor’ had been away from the family.” Saccades
are the thin lines; fixations are the knot-like interruptions.

Links on yarbus experiment

* http://viewonperception.wordpress.com/page/2/

* http://sstetson.wordpress.com/2010/07/
   quote from Yarbus:
	“Depending on the task in which a person is engaged, the distribution
   	of points of fixation on an object will vary correspondingly”
   	(Yarbus, 1967, p. 192)

* Marianne DeAngelusa & Jeff B. Pelza,
  Top-down control of eye movements: Yarbus revisited
  Visual Cognition, v.17: p.790-811, 2009
  DOI:10.1080/13506280902793843

  The Yarbus data was based on only a single observer, who wore a complex
  apparatus and had his head immobilized.
  From Abstract:

  Although Yarbus’ work is often cited to demonstrate the task-dependence of
  eye movements, it is often misrepresented; Yarbus reported results for only
  one observer, but authors commonly refer to Yarbus’ “observers”.
  Additionally, his observer viewed the painting for 21 minutes with optical
  stalks attached to the sclera and with his head severely
  restricted. Although eye movements are undoubtedly influenced by high-level
  tasks, it is not clear how Yarbus’ results reflect his unique experimental
  conditions. ... We replicated Yarbus’ experiment using a head-free
  eyetracker with 17 naïve observers. The presentations were self-paced;
  viewing times were typically an order of magnitude shorter than the times
  Yarbus imposed. Eye movement patterns were clearly task dependent, but some
  of the differences were much less dramatic than those shown in Yarbus’
  now-classic observations.

The block copying task of Dana Ballard: Two useful maxims

One of the first detailed studies of eye movements in relation to
manipulative activity was by Ballard, Hayhoe, Li, and Whitehead (1992). They
used a task in which a model consisting of coloured blocks had to be copied
using blocks from a separate pool. Thus the task involved a repeated sequence
of looking at the model, selecting a block, moving it to the copy and setting
it down in the right place (Figure). p.79

 
Figure: The block copying task of Ballard et al. (1992). A Copy (bottom
left) of the Model is assembled from randomly positioned blocks in the Source
area. Typical movements of hand and eye are shown, together with their timing
in a typical cycle. The eyes not only direct the hands, but also perform
checks on the Model to determine the colour and location of the block being
copied.

The most important finding was that the operation proceeds in a series of
elementary acts involving eye and hand, with minimal use of memory. Thus a
typical repeat unit would be as follows: Fixate (block in model area);
remember (its colour); fixate (a block in source area of the same colour);
pickup (fixated block); fixate (same block in model area); remember (its
relative location); fixate (corresponding location in model area); move
block; drop block. The eyes have two quite different functions in this
sequence: to guide the hand in lifting and dropping the block, and,
alternating with this, to gather the information required for copying (the
avoidance of memory use is shown by the fact that separate glances are used
to determine the colour and location of the model block). The only times that
gaze and hand coincide are during the periods of about half a second before
picking up and setting down the block (as with other tasks the eyes have
usually moved on before the pickup or drop are complete).

The main conclusion from this study is that the eyes look directly at the
objects they are engaged with, which in a task of this complexity means that
a great many eye movements are required. Given the relatively small angular
size of the task arena, why do the eyes need to move so much? Could they not
direct activity from a single central location? Ballard et al. (1992) found
that subjects could complete the task successfully when holding their gaze on
a central fixation spot, but it took three times as long as when normal eye
movements were permitted.

For whatever reasons, this strategy of “do it where I’m looking” is crucial
for the fast and economical execution of the task. This strategy seems to
apply universally. With respect to the relative timing of fixations and
actions, Ballard, Hayhoe, and Pelz (1995) came up with a second maxim: the
“just in time” strategy. In other words the fixation that provides the
information for a particular action immediately precedes that action; in many
cases the act itself may occur, or certainly be initiated, within the
lifetime of a single fixation. It seems that memory is used as little as
possible.