Content uploaded by Kieran Mohr
Author content
All content in this area was uploaded by Kieran Mohr on Apr 09, 2020
Content may be subject to copyright.
Modulation of the C1 component of the visual evoked potential
by spatial attention
Kieran S. Mohr, Niamh Carr, Simon P. Kelly, University College Dublin
The authors would like to thank both the Irish Research Council and Science Foundation Ireland for funding this work
and would like to further extend their thanks to Rachel Georgel for her assistance in data collection for experiment 3.
Project Code: 15/CDA/3591
Project Code: GOIPG/2016/123
V1
V2
V2
Informs Stimulus
Positions
Multifocal Mapping (Before Experiment)
Typically, the initial afferent activity of V1 has been measured in VEPs via the C1 component, which peaks 70-90 ms
post-stimulus and is canonically associated with a V1 source due to its polarity reversal across the horizontal
meridian, reflecting the opposing faces of the Calcarine banks along which V1 lies. Below, we demonstrate a
related signal resulting from pattern-pulse multifocal stimulation whose pattern of topography as stimuli are
presented around the visual field is highly consistent with a source of V1’s morphology. Thus, we used this
stimulation protocol to optimally choose stimulus locations for individual participants prior to conducting the
spatial attention experiments, due to the likely dominance of V1 in the signal and the speed with which the signal
can be recorded (10 minutes of stimulation time).
Experiments
Motivation: Although a previous study found strong attentional modulation of the C1[1] in a target detection
paradigm, two recent replication attempts failed to reproduce the result[2,3].
We have proposed two possible explanations for this discrepancy[4]:
1. The task was to detect the presence/absence of a ring-shaped luminance reduction (low spatial frequency)
atop a Gabor patch (high spatial frequency). In the original experiment only, the Gabor patch had a net
luminance brighter than the background, which introduced a low spatial frequency component. Thus, low
spatial frequencies did not uniquely indicate a target. However, the replication attempts used pure-
contrast Gabor patches and so low spatial frequencies could uniquely indicate a target. Therefore,
attentional modulation may have been limited to low spatial frequency coding neurons and thus may have
left the Gabor patches unmodulated.
2. Participants moved among different difficulty levels based on their performance, however only in the
original experiment were they informed of their difficulty level on an ongoing basis. This may have
facilitated a greater level of learning or indeed a higher level of motivation.
Experiment 1 (N=15): Compare two target-type conditions:
an orthogonal Gabor and a uniform disc shaped luminance increment
Experiment 2 (N=15): Compare thorough feedback and minimal feedback conditions (Thorough feedback
condition always first)
Experiment 3 (N=15): Close repetition of the original experiment with the darkened background and ring-
shaped targets
Stimuli
Experiments 1+2
Orthogonal Gabor
Experiment 1
Uniform Disc
Experiment 3
Ring
(A) Difficulty levels of the three different target
types. (B) Stimuli used in the two failed
replications[2,3] of [1]. The black, red and blue bars
indicate cross sections of the stimulus that are
plotted below in panel D. The adjacent heat maps
are 2-D Fourier transforms of these stimuli showing
the low and high spatial frequency components. (C)
Similar to Bexcept for stimuli used in experiment 3
and in [1]. (D) Cross sections of the pure contrast
Gabor patches of the failed replications (left) and
the net luminance increment Gabor patches of [1]
and experiment 3 (right). The luminance
decrements of the ring are shown on bottom.
Cycles/degree
Cycles/degree
Cycles/degree Cycles/degree
FFT Amplitude FFT Amplitude
Cycles/degree
Cycles/degree
FFT Amplitude
Cycles/degree
Cycles/degree
FFT Amplitude
Failed
Replications
Original
Stimulus
C1 Across All Experiments
Collapsing across all three experiments, we found that the C1 was significantly larger
in the attended than the unattended condition (p=.001) with a medium effect size
(ηp2=0.24)
C1 & P1 (Individual Experiments) Behaviour
Background
The C1 component of the visual evoked potential (VEP) represents the earliest latency at which we can
observe visual processing on the human scalp. While subsequent VEP components are routinely found to be
modulated by spatial attention, such modulations of the C1 component have been rarely observed.
C1 amplitude was significantly higher in the attended condition for experiments 2
(p<.05) and 3 (p<.01). In experiment 1, C1 amplitude was instead reduced by
attention but only in the upper visual field without CSD transformation (p<.01)
Between experiments 1 and 2, the patterns of modulation of both the C1 and the
contralateral P1 were quite different. While experiment 2 yielded enhancements of
both the C1 and the P1 (p<.001), experiment 1 yielded a reduced amplitude of the
C1 in the upper field and no modulation of the contralateral P1 in the lower field
(p>.1), although this modulation was present in the upper field (p<.001).
Furthermore, the P1 began earlier in experiment 1 than in experiment 2.
Higher performance levels
were achieved in experiment
1 than in experiment 2. In
experiment 1, participants
reached higher difficulty
levels (p<.05) and were more
accurate at the high difficulty
levels (p<.01) than were
participants in experiment 2.
This is despite restricting the
comparison to the conditions
in each experiment that were
identical to one another (the
Gabor-target condition of
experiment 1 and the high
feedback condition of
experiment 2).
Discussion
By collapsing data across all three experiments (N=45)we could conduct a highly powered test of modulation of the C1 component of
the VEP by spatial attention, demonstrating a significant modulation of medium effect size.
Contrary to this overarching trend, experiment 1 exhibited a reduction of C1 amplitude by attention in the upper visual field, an
effect that was eliminated by CSD transformation. This same experiment also exhibited an elimination of the P1 attention effect in
the lower visual field. This stood in stark contrast to the identical condition of experiment 2 where the results were in line with
expectation.
These perplexing results may be explained by signal overlap. While the C1 is generally thought to reflect mostly V1 activity,
contributions from V2 and V3 are also likely[5]. Due to cortical geometry, V2/V3 sources oppose V1 sources on the scalp but P1
polarity congruency with these signals differs between the upper and lower visual field (congruent with V1 in the lower and with
V2/V3 in the upper visual field). Thus, a difference in strategy between the experiments that produced greater modulation of V2/V3
in experiment 1 could potentially explain these results. Although unintended, the orthogonal grating target may have injected subtle
texture information into the stimulus compound. Thus, such a strategy could be consistent with greater sensitivity of V2 to the
detection of texture[6,7].
5. Ales, J. M., Yates, J. L., & Norcia, A. M. (2013). On determining the intracranial sources of visual evoked potentials
from scalp topography: A reply to Kelly et al.(this issue). NeuroImage, 64, 703-711.
1. Kelly, S. P., Gomez-Ramirez, M., & Foxe, J. J. (2008). Spatial attention modulates initial afferent activity in human
primary visual cortex. Cerebral cortex,18(11), 2629-2636.
2. Baumgartner, H. M., Graulty, C. J., Hillyard, S. A., & Pitts, M. A. (2018). Does spatial attention modulate the earliest
component of the visual evoked potential?. Cognitive neuroscience,9(1-2), 4-19.
3. Alilović, J., Timmermans, B., Reteig, L. C., Van Gaal, S., & Slagter, H. A. (2019). No evidence that predictions and
attention modulate the first feedforward sweep of cortical information processing. Cerebral Cortex,29(5), 2261-2278.
4. Kelly, S. P., & Mohr, K. S. (2018). Task dependence of early attention modulation: the plot thickens. Cognitive
neuroscience,9(1-2), 24-26.
References
Topographies
C1b (Opposite Polarity of C1)
Lower Visual Field
CSD Transformed
Since the grand average topographies demonstrated the
emergence of a midline signal of negative polarity appearing
for the lower visual field under CSD transformation only, we
measured this signal as a possible index of V2/V3 activity.
The signal itself was weak (ηp2=0.29 for signal presence vs
ηp2=0.86 for the C1) so we restricted the analysis to data
collapsed across all three experiments. We found that this
signal was marginally enhanced by attention (p=.049).
6. Ziemba, C. M., Freeman, J., Movshon, J. A., & Simoncelli, E. P. (2016). Selectivity and tolerance for visual texture
in macaque V2. Proceedings of the National Academy of Sciences,113(22), E3140-E3149.
7. Merigan, W. H., Nealey, T. A., & Maunsell, J. H. (1993). Visual effects of lesions of cortical area V2 in
macaques. Journal of Neuroscience,13(7), 3180-3191.
