Project Narrative
Schizophrenia has heterogeneous effects on brain function. The effects on the
smooth pursuit and saccade systems are well documented. The effect of
schizophrenia on other classes of eye movement has not been explored. This study
will use binocular eye tracking and real stereo targets to investigate abnormalities
of vergence eye movements in schizophrenic subjects.
Neural substrates of vergence eye movements.
Bold: Areas containing cells that are related to vergence eye movements.
Italics: Areas likely to contain vergence-related cells based on anatomical and
electrophysiological studies.
Unidentified areas are shown by question marks.
EW= nucleus of Edinger-Westphal; F= fastigial nucleus; FEF= frontal eye fields; IP=
posterior interposed nucleus; NRTP= nucleus reticularis tegmenti pontis; SOA=
supra-oculomotor area) (Adapted from Gamlin, 2002 and Gurler 2007).
Cortical areas involved in control of vergence eye movements in humans. FEF=
frontal eye fields; SEF= supplementary eye fields, MT= middle temporal area; MST=
medial superior temporal area, PPC= posterior parietal cortex, PEF= parietal eye
fields (Adapted from Bϋttner & Bϋttner-Ennever, 2006).
Specific Aim 1 will be to replicate the results of previous studies in a set of
schizophrenic volunteers. This will validate the experimental setup and provide
data for comparison and contrast with the results in the following aim. In these
experiments smooth pursuit gain; intrusive saccades during smooth pursuit and
error rate in an antisaccade task will be measured in each subject. Based on
previous reports smooth pursuit is expected to have lower gain and an increased rate
of intrusive saccades when compared to normal volunteers. Antisaccade error rate is
expected to be higher than that found in normal volunteers. These results from these
experiments will be used in aim 3 to generate combined measures of abnormality.
The relationship of these anomalies to those measured in aim 1 may suggest
something about the mechanism or underlying deficits in the eye movements of
schizophrenics.
Group Runs Mean Std Dev
Schizophrenia 26 114.5 29.1
SPEM
Schizophrenia 9 59 13.7
Vergence
Tracking
Control SPEM 9 78.6 8.2
Control 4 48.3 19.1
Vergence
Tracking
Pooled 48 91.8 23.5
Bartlett's statistic 14.9815
Degrees of freedom 3
p-value 0.0018
Count Mean Std Dev
sz spem 26 114.5 29.0892
sz vt 9 59 13.6931
nv spem 9 78.556 8.2327
nv vt 4 48.25 19.1377
Pooled 48 91.833 23.4983
Bartlett's statistic 14.9815
Degrees of freedom 3
p-value 0.0018
predictive vergence
A big part of this depends vergence results that I do not have yet…
Look at data. Look at data again. Basic stats and sanity checks (no pun
intended) and some data reduction and clustering. Try Ggobi?
A technique like Logistic regression, LDA, or probit analysis in the binary
classifier model (mapping onto schizophrenia DSM-IV diagnosis) or Regression
analysis to map measures onto continous instruments like BPRS. Use results to
see which measures are fundamental. Take most important measures, pick a
sensitivity for classifier, estimate best threshold for measures. Report ROC.
Failing that, try non-linear then non-parametric regression… then generate
classifier.
Failing that fall back to some variety of machine learning algorithm like SVM
or NN or something.
Test results with bootstrap scheme.
Or something like that.
Eye movements are also important in non-verbal communication. A fact that
becomes quite apparent when interacting with individuals with eye movement
disorders.
Dodge in 1903, described five kinds of eye movement in the horizontal plane. 5
Three kinds locate and keep targets on the fovea and two to stabilize the view
during motion of the head.
The first kind of eye movement Dodge defined were saccades. These are
ballistic eye movements that jerk the eye and rapidly move the fovea from target
to target in the visual scene. Watching someone read provides a clear
demonstration of these movements. Vision is attenuated during these
movements, which is demonstrated by the inability to see ones own saccades in
a mirror.
The second kind were smooth pursuit. This fluid eye motion allows the fovea to
track small targets moving across a stationary background. This movement is
demonstrated by watching someone track a fingertip or the end of a pen as it
moves slowly across the field of view. This motion is voluntary but it requires a
target to initiate. This can be seen by asking someone to move his or her eyes
smoothly without a target and observing the resulting saccades.
The fourth kind were compensatory reflexive eye movement that stabilize the
visual scene on the retina during motions of the head. General movement of the
visual scene will drive these movements and they are also driven by the
vestibular and proprioceptive systems. These movements are further defined
below. However, two common examples are: watching a subway train pull out of
a station resulting in your eyes being “pulled” along by the train, and the ability to
read a text like this one while nodding or turning your head from side to side. You
should be nodding now.
The fifth kind of eye movement defined by Dodge is the jerky, reflexive,
compensatory eye movements called nystagamus. These are eye movements
that reset the eye position after it has turned through a large angle. These
movements do not cease immediately after prolonged rotation, so they are easily
observed by spinning someone in a chair many times and then stopping the
rotation. They can be felt after prolonged self-rotation by closing your eyes and
gently touching your fingertips to your eyelids.
The fifth and final kind of eye movements described by Dodge is disconjugate
and referred to as vergence eye movements. Unlike the conjugate movements
above, the eyes move in opposite directions. Convergence is inward turning of
the eyes to view near objects and divergence is outward turning of the eyes to
view objects at a distance. These movements can be observed by asking
someone to look at the tip of your finger while you move your finger closer and
farther from their eyes. Vergence eye movements align the foveae of both eyes
on the same target in space. Objects that are nearer or farther than the target
may appear as a double image. These objects are out of the plane of regard and
the image in the non-dominant eye may be suppressed.
Much work has been done since the seminal investigations of Dodge.
(Smyrnis 2008; Rommelse et al. 2008; Turetsky et al. 2007; Reuter & Kathmann
2004; Trillenberg et al. 2004; McDowell & Clementz 2001; Holzman 2000; Hutton
& Kennard 1998; Levy et al. 1994; Abel et al. 1992; Gaebel 1989; Holzman 1985)