Pergamon
PIh S0042-6989(96)00317-3
Vision Res., Vol. 37, No. 13, pp. 1845-1849, 1997
© 1997 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
0042-6989/97 $17.00 + 0.00
Contrast Dependency of Motion-onset and
Pattern-Reversal VEPs: Interaction of Stimulus
Type, Recording Site and Response Component
MICHAEL BACH,*]" DIETER ULLRICH*
Received 1 December 1995; in revised form 29 August 1996
We compared the contrast dependency (from 0.4 to 98%) of the visual evoked potential (VEP) to
motion onset and to pattern reversal at an occipital and lateral recording site using sinewave
grating stimuli of 0.9 c/deg, drifting at 4.9 deg/sec. Two differing VEP components were identified:
a positive component, peaking at around 130 msec, dominating the occipital derivation, enhanced
in pattern-reversal stimulation, exhibiting a high-threshold, late-saturating contrast response
characteristic with a half-amplitude contrast above 7%; and a negative component at around
180 msec, dominating the lateral derivation, enhanced in motion-onset stimulation, exhibiting a
low-threshold, saturating contrast characteristic with a half-amplitude contrast below 4%. The
results suggest: (1) The negative component (N180) represents motion mechanisms, located more
laterally, while the positive component (P100-P130) represents form-processing mechanisms,
located near the V1/V2 areas. (2) A pattern-reversal stimulus triggers both form-processing and
motion mechanisms that can be discriminated by latency. In an occipital derivation, the clinical
reversal VEP PI00 will be little contaminated by motion responses. © 1997 Elsevier Science Ltd.
Human Motion-onset Pattern-reversal Contrast Visuallyevoked potentials
INTRODUCTION
Different qualities of vision seem to be processed to some
degree in different pathways (e.g., Lennie et al., 1990;
Casagrande, 1994). Motion processing mechanisms carry
the signature "low contrast threshold with a saturating
response characteristic" throughout the visual pathways,
as is suggested by evidence from various fields: (1) the
M-pathway has a lower contrast threshold than the
P-pathway (Kaplan & Shapley, 1986; Sclar et al., 1990);
(2) motion perception is related to the target area of the
M-pathway (e.g., Maunsell et al., 1990); (3) contrast gain
functions of moving gratings, measured by functional
MRI, show a low contrast saturating response characteristic in area MT (Tootell et al., 1995); (4) motion
perception has a high contrast sensitivity [e.g., Keck et al.
(1976) found no dependency of the motion aftereffect on
adapting contrast, provided that the adapting contrast was
above 3%].
Measurement of electrophysiological correlates of
motion processing in man was pioneered 20 years ago
by Clarke (1973) and has recently received renewed
interest (e.g., Tyler & Kaitz, 1977; GOpfert et al., 1990;
*Universit/its-Augenklinik, Killianstr. 5, D-79106 Freiburg, Germany.
~'To whom all correspondence should be addressed [Tel +49-(761)
270-4060/4061; Fax +49-(761) 270-4052; Email [email protected]].
Manning et al., 1991; Norcia et al., 1991; Ullrich & Bach,
1992; Wesemann & Norcia, 1992; Kommerell et al.,
1995; Kubovfi et al., 1996; Snowden et al., 1995), driven
both by the popular magno/parvo distinction and
advancements in stimulation technology. Disagreement
on the motion-specific component (positivity at 130 msec
or negativity at around 180 msec) has been resolved and
traced to adaptation by the test stimulus (Bach & Ullrich,
1994). Not every moving stimulus, however, evokes a
motion-specific VEP (Snowden et al., 1995). The abovementioned "motion signature" was employed in a recent
study comparing motion-onset and pattern-reversal
(Kubovfi et al., 1995) with the conclusion that the negative component of the pattern-reversal VEP at around
180 msec was related to motion. In their amplitude vs
contrast functions (their Fig. 4), these authors did not
discriminate between the recording sites, they took the
maximal response from the various derivations. Furthermore, they employed a moving mirror for patternreversal. Thus, their reversal stimulus was based on
physical motion, though rapid, risking the possibility of a
"true motion" artifact in the stimulus that was compared
to a motion stimulus.
As pattern reversal is the most common stimulus in
clinical settings (Harding et al., 1996), possible confounding of motion and contrast mechanisms should be
clearly assessed. We therefore report on an experiment
similar to the one conducted by Kubovfi et al. (1995),
1845
1846
M. BACH and D. ULLRICH
using both motion-onset and "true" pattern-reversal
stimuli on a visual display unit and analysing the effects
at each recording site separately.
METHODS
Subjects
Eight subjects (aged 25-35 yr) participated in the
experiments. They had a corrected visual acuity of _>1.0.
The only selection criterion was normal acuity and
normal amplitude (_>5 gV) in a clinical standard chequerboard reversal VEP.
Stimuli
Stimuli were created with a Cambridge Research
VSG2 graphics card and displayed on an Eizo 9070S
raster-scan visual display unit with a frame rate of 71 Hz.
For both motion and reversal stimulation we used
sinusoidal gratings with a spatial freq,uency of 0.9
cycles/deg, mean luminance of 15 cd/m ~ and contrast
from 0.4 to 98% in eight steps. The circular stimulation
area with a diameter of 11 deg was surrounded by a grey
mask with a mean luminance of 17 cd/m 2. A motion
sequence consisted of an abrupt onset of continuous
motion at 4.9 deg/sec for 140msec, followed by a
stationary phase for 1260 msec. This resulted in a duty
cycle of 10% where little adaptation occurs (Bach &
Ullrich, 1994). For pattern-reversal stimuli, rapid phase
reversal occurred every 900 msec, corresponding to
1.1 rev/sec. Luminance and contrast of the monitor were
calibrated employing a method similar to that described
by Pelli and Zhang (1991).
Recording
VEP was recorded from an Oz-Fpz derivation and a
lateral (5 cm left of Oz vs. right ear) derivation using
gold-cup electrodes. The ground electrode was attached
to the right wrist. Signals were amplified and filtered
(first-order band-pass, 0.5-70 Hz, Toennies Physiological Amplifier) and digitized to a resolution of 12 bits at a
sampling interval of 2.33 msec. The computer averaged
the sweeps if their amplitude did not exceed + 100 ffV,
Motion onset
Reversal
Contrast
100%
60%
20%
Oz-Fpz
14%
6.0%
4.0%
1.5%
>
0.4%
'
. . . . . . . .
9(~0
0
<E
700
Time [ms]
I O0
100%
60%
Left
lateral
20%
14%
VS.
6.0%
right ear
4.0%
1.5%
0.4%
o~ . . . . . . . .
9ool
o~ . . . . . .
7oo~. . . . . .
1~oo
FIGURE 1. Grand mean responses (bold lines; + SEM thin lines) of eight subjects to motion-onset/offset stimuli. The thin
vertical lines at 140 msec represent the offset of motion.
CONTRAST EFFECTS ON MOTION AND REVERSAL VEPS
displayed them on-line and generated the stimuli.
Averaged sweeps were digitally filtered with a bandpass from 1 to 30 Hz. Amplitudes were measured off-line
from baseline to peak.
Procedure
At the beginning of each session we measured VEP
responses to luminance-contrast chequerboard patterns as
described in the ISCEV VEP standard (Harding et al.,
1996) to ascertain correct operation of the entire set-up.
We then presented the various stimuli in a counterbalanced interleaved block design: each stimulus appeared five times, then the next stimulus followed. This
sequence was repeated 40 times, resulting in a total of
200 sweeps for each condition. After every 50 sweeps,
resting periods of 2 min duration were interposed. The
entire recording session lasted about 2 hr. The subjects
were requested to fixate a cross at the centre of the screen
and report random digits that appeared there for 300 msec
at random intervals.
RESULTS
Figure 1 displays the grand average across all
subjects_+ SEM for the two stimuli (reversal on the left,
and motion onset right), all contrast levels (increasing
from bottom to top), and the two recording sites (Oz vs
Fpz, top; left lateral vs right ear, bottom). As a common
characteristic for both stimulus types and recording sites
we observed two major components, a positive one at
around 130 msec and a negative one at around 180 msec,
appearing at a contrast of 1.5% and above. At the lowest
contrast level of 0.4%, no response could reliably be
identified.
The contrast dependency of the two components is
depicted in Fig. 2. To avoid clutter of the graph, the
measures of error in Fig. 1 have been omitted in Fig. 2.
Inspired by Sclar et al. (1990), we fitted Naka-Rushton
equations to the amplitude vs log contrast values with
three free parameters:
Cn
A(c) ----Amax •
o 7
5
4
2"
2
~ / _
1
J
0
- ~- - . ~ r ~ .
O
J..C~ "'1~O .....
2%
t-~,~-O
.................
1
10
1
0
100
1
1o
1oo
7-
7-
+
o ---
6-
Positivity
NegatMty
65-
5-
4-
0.95%
/
,~"
/
2-
¢~
~ ' ~ - "
0 ~*
- I
I
Left lateral
~,t~.,-
o
~
.
+
~
+
. . . . . . . . . . . .
I0
0
0
VS.
right ear
1o%
~. ooooooO'*°'*
~ ......
100
.---O"'13
.o~°
3-
~
e"
2.0%
44.2%
3-
1-
O z vs. Fpz
7.0%
0.95%
3.
c" + C~o
7.
4-
E
<
-
Motion On
+
5.
-
where A = VEP amplitude depending on contrast c,
Reversal
>
1847
,+
1
,~,?
....
~-!
,..!.,~-.,
10
100
Contrast
[%]
FIGURE 2. Amplitude vs log contrast for the positive component (P130, plus symbols) and the negative component (N180,
circles), for the two stimulus types (reversal left, motion right) and the two recording sites (occipital top, lateral bottom). The
continuous curves represent Naka-Rushton fits, the numbers at the arrows represent the half-amplitude contrast. The positive
component displays an accelerating contrast characteristic, and the negative component a saturating type. In the occipital lead
for the reversal stimulus the negative component is very low and least resembles a saturating response characteristic, the
positive component dominates. In the lateral lead for the motion onset stimulus the positive component is small. The negative
component dominates, displaying a low contrast threshold.
1848
M. BACH and D. ULLRICH
Ar,~× = maximal amplitude, C~o = contrast for 50% Amax
and n = exponent, representing the steepness of rise
[constraint: n _< 3].
As the positive component did not enter the saturating
region, c~o often lies above 100%. Still, the general shape
of the response characteristic seems well represented, and
two qualitatively different types emerge. As a single
distinguishing measure, we computed the contrast where
the VEP amplitude is halfway between the maximal and
minimal level (denoted as "half-amplitude contrast",
arrows in Fig. 2).
The most obvious features were a non-saturating
response function to pattern reversal in the occipital
positive response component (top left, marked with plus
signs) and a saturating response function to motion onset
in the lateral negative response component (bottom right,
marked with circles).
Comparing the positive and negative components
across stimulus types and recording sites, we observed
that the positive component displayed a non- (or late-)
saturating function characteristic. Its half-amplitude
contrast was always >_7%. The negative component
displayed a low-threshold, saturating type of response
function with half-amplitude contrast always below 4%.
Comparing the effect of recording site across stimuli,
we found that in the lateral derivation the negative
component clearly dominated for both stimulus types; in
the occipital derivation, the positivity was relatively
enhanced. Comparing the effect of stimulus type across
recording sites, we found that the motion-onset stimulus
enhanced the negative component, and pattern reversal
enhanced the positive component.
DISCUSSION
Well-defined VEP responses were identified for both
motion-onset and pattern-reversal stimulation for a
contrast of 1.5% and above. Two components dominated
the response, a P130 and a N180. Going from reversal to
motion stimulation, and from an occipital to a lateral
recording site, we observed a transition from a nonsaturating positivity-dominated response to a low-threshold, saturating negativity-dominated response characteristic (Fig. 2).
The finding of a high contrast sensitivity of the
negative component agrees well with work by Mtiller
and G6pfert (1988), who were the first to report on the
high contrast sensitivity of the motion-onset VEP, and
with Kubov~ et al. (1995). There is further evidence from
motion adaptation experiments (Bach & Ullrich, 1994)
that the negative component (N180) represents motion
processing as originally suggested by Clarke (1973) and
G6pfert et al. (1984). In the light of the "motion
signature" (see Introduction), the high contrast sensitivity
of the negative component strengthens the evidence for
the N180 as a "motion" response, as suggested by Kuba
& Kubov~ (1992). While our own findings would be only
based on the specific situation of sinusoidal gratings with
the specified temporal and spatial properties, the
similarity to Kubov~ and colleagues' (Kubov~ et al.,
1995) chequerboard findings suggests that this is a fairly
general conclusion.
The lateral dominance of the motion response agrees
well with results by Probst et al. (1993), who used
random-dot motion stimulation and advanced source
location and found the generator of the motion responses
to be located in the lateral region, possibly homologous to
the area MT or MST in non-human primates.
Pooling the contrast dependency across recording sites,
the present findings agree closely with the one reported
by Kubovfi et al. (1995), suggesting that motion artifacts
in the reversal condition have not played an appreciable
role in that experiment. The present results extend those
findings to lower contrast levels; their lowest contrast was
1.3%, our lowest level was 0.4%. Figure 2 shows that
there is a fairly sharp drop for the motion response below
1.5%. With respect to recording site the present findings
extend those by Kubov~ et al. (1995) who reported
always the largest amplitude, irrespective of site. The
reversal stimulus produces a negative component at the
occipital derivation which is fairly small compared to the
one evoked by motion onset and does not display the
typical early saturating response type.
The present results are also of interest from a clinical
perspective: the positive component at 130 msec most
probably corresponds to the well-known P100; latency to
sinusoidal stimulation is higher as compared to squarewave gratings or chequerboards (Bobak et al., 1987).
This component does not display the "motion signature".
On the other hand, the response will be dominated by
motion processing mechanisms under conditions of low
contrast (e.g., 10%), low duty cycle (10% to avoid motion
adaptation) and a lateral derivation (Kuba & Kubov~,
1992). We conclude that a typical P100 response to a
high-contrast pattern-reversal stimulus, recorded from
occipital leads, contains little contamination from motion
responses.
REFERENCES
Bach, M. & Ullrich, D. (1994). Motion adaptation governs the shape of
motion-evoked cortical potentials (motion VEP). Vision Research,
34, 1541-1547.
Bobak, P., Bodis-Wollner, I. & Guillory, S. (1987). The cffcct of blur
and contrast on VEP latency: comparison between check and
sinusoidal grating patterns. Electroencephalography & Clinical
Neurology, 68, 247-255.
Casagrande, V. A. (1994). A third parallel visual pathway to primate
area V1. Trends in Neuroscience, 17, 305-310.
Clarke, P. G. H. (1973). Comparison of visual evoked potentials to
stationary and moving patterns. Experimental Brain Research, 18,
156-164.
G6pfert, E., Mtiller, R. & Simon, E.-M. (1990). The human motion
onset VEP as a function of stimulation area for foveal and peripheral
vision. Documenta Ophthalmologica, 75, 165-173.
G6pfert, E., Mtiller, R. & Hartwig, M. (1984). Effects of movement
adapation on movement-visual evoked potentials. Documenta
Ophthalmologica Proceedings Series, 40, 321 324.
Harding, G. F. A., Odan, J. V., Spileers, W. & Spekreijse, H. (1996).
Standard for visual ew)ked potentials. Vision Research, 36, 35673572.
Kaplan, E. & Shapley, R. M. (1986). The primate retina contains two
types of ganglion cells, with high and low contrast sensitivity.
CONTRAST EFFECTS ON MOTION AND REVERSAL VEPS
Proceedings of the National Academy of Sciences USA, 83, 27552757.
Keck, M. J., Palella, T. D. & Pantle, A. (1976). Motion aftereffect as a
function of the contrast of sinuisodal gratings. Vision Research, 16,
187-191.
Kommerell, G., Ullrich, D., Gilles, U. & Bach, M. (1995). Asymmetry
of motion VEP in infantile strabismus and in central vestibular
nystagmus. Documenta Ophthalmologica, 89, 373-381.
Kuba, M. & Kubov~, Z. (1992). Visual evoked potentials specific for
motion onset. Documenta Ophthalmologica, 80, 83-93.
Kubov~, Z., Kuba, M., Juran, J. & Blakemore, C. (1996). Is the motion
system relatively spared in amblyopia? Evidence from cortical
evoked responses. Vision Research, 36, 181-190.
Kubovfi, Z., Kuba, M., Spekreijse, H. & Blakemore, C. (1995).
Contrast dependence of motion-onset and pattern-reversal evoked
potentials. Vision Research, 35, 197-205.
Lennie, P., Trewarthen, C., Van Essen, D. & W~issle, H. (1990).
Parallel processing of visual information. In Spillmann, L. &
Werner, J. S. (Eds), Visual perception. The neurophysiological
foundations (pp. 103-128). San Diego: Academic Press.
Manning, M. L., Finlay, D. C. & Fulham, W. R. (1991). Time-tillbreakdown and VEP measures of short-range apparent motion.
Vision Research, 31, 1865-1874.
Maunsell, J. H., Nealey, T. A. & De Priest, D. D. (1990). Magnocellular and parvocellular contributions to responses in the middle
temporal visual area (MT) of the macaque monkey. Journal of
Neuroscience, 10, 3323-3334.
Mfiller, R. & G6pfert, E. (1988). The influence of grating contrast on
the human cortical potential visually evoked by motion. Acta
Neurobiology Experimenta, 48, 239-249.
Norcia, A. M., Garcia, H., Humphry, R., Holmes, A., Hamer, R. D. &
Orel-Bixler, D. (1991). Anomalous motion VEPs in infants and in
infantile esotropia. Investigative Ophthalmology and Visual Science,
32, 436-439.
1849
Pelli, D. & Zhang, L. (1991). Accurate control of contrast on
microcomputer displays. Vision Research, 31, 1337-1350.
Probst, T., Plendel, H., Paulus, W., Wist, E. R. & Scherg, M. (1993).
Identification of the visual motion area (area V5) in the human brain
by dipole source analysis. Experimental Brain Research, 93, 345351.
Sclar, G., Maunsell, J. H. R. & Lennie, P. (1990). Coding of image
contrast in central visual pathways of the macaque monkey. Vision
Research, 30, 1-10.
Snowden, R. W., Ullrich, D. & Bach, M. (1995). Isolation and
characteristics of a steady-state visually evoked potential in humans
related to the motion of a stimulus. Vision Research, 35, 1365-1373.
Tootell, R. B. H., Reppas, J. B., Kwong, K. K., Malach, R., Born, R. T.,
Brady, T. J., Rosen, B. R., Belliveau, J. W. (1995). Functional
analysis of human MT and related visual cortical areas using
magnetic resonance imaging. Journal of Neuroscience, 15, 32153230.
Tyler, C. W. & Kaitz, M. (1977). Movement adaptation in the visual
evoked response. Experimental Brain Research, 27, 203-209.
Ullrich, D. & Bach, M. (1992). Components related to motion onset/
offset in visually evoked potentials. Perception, 21, 32.
Wesemann, W. & Norcia, A. M. (1992). Contrast dependence of the
oscillatory motion threshold across the visual field. Journal of the
Optical Society of America, 9, 1663-1671.
Acknowledgements--This work was supported by the Deutsche
Forschungsgemeinschaft (SFB 325, B3) and the Meyer-SchwarttingStiftung. We thank S. Waltenspiel, H. Kalmbach and T. Meigen for
help in programming, Bob Snowden, Tony Norcia and unknown
reviewers for critical comments on the manuscript and our subjects for
their patience.