Psychophysical channels and ERP population responses in human visual cortex: Area summation across chromatic and achromatic pathways

Vision Research 50 (2010) 1283–1291

Contents lists available at ScienceDirect

Vision Research journal homepage: www.elsevier.com/locate/visres

Psychophysical channels and ERP population responses in human visual cortex: Area summation across chromatic and achromatic pathways Maria J. Ribeiro *, Miguel Castelo-Branco IBILI, Faculty of Medicine, University of Coimbra, Portugal

a r t i c l e

i n f o

Article history: Received 19 November 2009 Received in revised form 20 April 2010

Keywords: Visual evoked potentials Color vision Contrast gain Area summation Human

a b s t r a c t In the early stages of vision, information is transmitted through distinct physiologically defined pathways. These may be related with three post-receptoral detection mechanisms defined psychophysically in humans. Accordingly, the parvocellular pathway is very sensitive to L-M-cone contrast, processes mainly foveal information and underlies fine discrimination of visual features. The magnocellular pathway is most sensitive to luminance contrast and is important for visuo-spatial and motion processing. The less understood koniocellular pathway responds to S-cone modulation outside the foveola. As such, the three pathways process visual information in a different manner, with the L-M-cone psychophysical channel being more devoted to central vision and the two other channels responding significantly to peripheral information. We measured size response functions of these three processing channels using event related potential (ERP/EEG) recordings and stimuli with various sizes and contrasts with the aim of studying coding of stimulus properties within each of these channels. The effect of stimulus size was significantly smaller for the L-M-cone channel consistent with its dominance in the central visual field. Furthermore, for this pathway, the effect of size was not modulated by stimulus contrast. In contrast, both the S-cone and achromatic channels showed a strong effect of size that was significantly modulated by contrast. Interestingly, both the S-cone and achromatic channels responded proportionally to the area of cortex activated, suggesting that the S-cone channel represents space in a similar manner to the achromatic channel. In conclusion, a fundamental relation exists between previously identified psychophysical mechanisms and population responses in the visual cortex. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Visual perception of low level features such as contrast and size emerges from population responses in early visual cortex (Boon, Suttle, & Henry, 2005; Boynton, Demb, Glover, & Heeger, 1999; Page & Crognale, 2005; Ress & Heeger, 2003). However, population responses are not easily determined from knowledge of single cell properties. In fact, in primary visual cortex, response dependence of single neurons on contrast varies considerably from cell to cell (Albrecht & Hamilton, 1982) and direct inference of population responses from single cell activity is not straightforward. To illustrate this point, Albrecht and Hamilton (1982) determined the contrastresponse functions of 247 neurons from striate cortex of monkey and cat. The dynamic range of the contrast-response functions of the neurons studied varied considerably so that ‘‘when considering the activation of a large population of cells, increasing the contrast of a grating produces an increase in both the overall number of action potentials produced, as well as, the overall number of cells responding” (Albrecht & Hamilton, 1982). Therefore, ‘‘there is no * Corresponding author. Fax: +351 239 480280. E-mail address: [email protected] (M.J. Ribeiro). 0042-6989/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.visres.2010.04.017

compelling a priori reason to expect [the population response] to fit any particular mathematical function”. Here we measured scalp responses related to population responses in human visual cortex, in order to probe simple mathematical relationships that attempted to model stimulus contrast and size response functions. The three post-receptoral detection mechanisms, defined psychophysically (Cole, Hine, & McIlhagga, 1993; Krauskopf, Williams, & Heeley, 1982; Sankeralli & Mullen, 1997), are thought to process visual information across space differently. These mechanisms have as their neural counterpart three retino-cortical pathways with their different physiological properties. The parvocellular pathway is thought to underlie fine discrimination of visual features in particular in the central visual field, and is sensitive to modulation of L-M-cone contrast, that results in red-green chromatic contrast. The magnocellular pathway is most sensitive to achromatic (luminance) contrast and is thought to play an important role in spatial localization and motion processing (Merigan & Maunsell, 1993). The function of the koniocellular pathway is less understood (Hendry & Reid, 2000). It most certainly plays a role in color vision due to its sensitivity to S-cone contrast. In addition, it might be involved in spatial processing, as suggested by its significant response to peripheral stimulation (Mullen, Dumoulin,

1284

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

McMahon, de Zubicaray, & Hess, 2007; Vanni, Henriksson, Viikari, & James, 2006), and in motion perception (Sincich, Park, Wohlgemuth, & Horton, 2004; Wandell et al., 1999). Using ERP/EEG recordings, we measured size response functions within each of these channels and their interaction with contrast. We were interested in determining how size response functions are affect by stimulus contrast and vice versa, because stimulus size influences contrast perception (Foley, Varadharajan, Koh, & Farias, 2007; Meese, Hess, & Williams, 2005; Wright & Johnston, 1982, 1983) and size response functions may be affected by stimulus contrast, just as, at the single cell level, spatial summation is strongly influenced by contrast (Solomon, Lee, & Sun, 2006). 2. Methods 2.1. Participants Fourteen healthy adults (9 female; average age 27; age range 22–41) volunteered for psychophysical and electrophysiological measurements after a full description of the aims and methods of this study. All subjects had normal or corrected to normal visual acuity and normal color vision as assessed with the Cambridge Color Test (Cambridge Research Systems, Rochester, UK). Informed consent was obtained from all participants. The study was conducted in accordance with the tenets of the Declaration of Helsinki, and with the approved guidelines of the Ethics Committee of the Faculty of Medicine of Coimbra. 2.2. Visual stimulation Stimuli were generated with MatLab (version 7.3.0.) and presented with the stimulation software Stim2 (Neuroscan) with a display resolution of 1024  768  32 and graphic processing unit NVIDIA GeForce 6600, provided by Neuroscan. The stimuli were presented in a CRT monitor (Philips) calibrated with a spectroradiometer (Spectrocal, Cambridge Research Systems). The spectrum of each phosphor was measured at 1 nm intervals across the visible spectrum. The Stockman and Sharpe (2000) 2-deg cone fundamentals were used for the spectral absorptions of the L-, M- and S-cones. From these data, a linear transform was calculated to specify the phosphor contrasts required for a given cone contrast (Brainard, Pelli, & Robson, 2002). Gamma correction of the monitor output was achieved via software look-up tables. The monitor refresh rate was set at 85 Hz. Stimuli were circular horizontal Gabors (sinewave gratings modulated by a Gaussian window) presented in phase reversal mode at the centre of the CRT monitor. The mean luminance of the stimuli and background was 35 cd/m2. Stimuli diameters, defined as two times the standard deviation of the Gaussian aperture filter, were 4.0°, 6.2°, 8.3° or 10.5° of visual angle. The viewing distance was 1 m and the screen subtended a visual angle of 13.7° in width and 18.2° in height.

Based on these cited reports, we chose the following cardinal directions: (1, 1, 1) – achromatic stimuli for isolation of the magnocellular pathway; (0, 0, 1) – S-cone modulating stimuli (S-cone stimuli) for isolation of the koniocellular pathway and (1, a, 0) – stimuli where L and M cones are modulated in antiphase (L-M-cone stimuli) for isolation of the parvocellular pathway. The variable ‘a’ represents the ratio L:M cone input into the luminance mechanism. This ratio can be determined for each subject by measuring the individual isoluminant point (L:M ratio at which the luminance mechanism is minimally activated). We determined this ratio by using a minimum motion technique (Cavanagh, MacLeod, & Anstis, 1987) for a patch of binocularly viewed L-M cone sinewave grating (2 cycles per degree). The average of the L:M ratios determined was 3.0, ranging from 2.1 to 4.9. These ratios are consistent with the values obtained by Mullen et al. (2007). The chromaticity of the L-M-cone stimuli was adjusted for each subject according to the determined isoluminant point. The Comission Internationale de l’Éclairage (CIE) x- and y-coordinates of our achromatic and S-cone stimuli were as follows: background and achromatic stimuli, x = 0.28, y = 0.30; +S-cone, x = 0.25, y = 0.21 and S-cone, x = 0.35, y = 0.53. The coordinates for the average L-M-cone stimuli, with L:M ratio of 3, were: L-Mcone, x = 0.33, y = 0.28 and M-L-cone, x = 0.23, y = 0.33. The luminance and chromatic modulation of the stimuli can be represented by two vectors of equal length symmetric about the origin of the cone contrast coordinates. The contrast at the centre of the Gabors was defined as the length of the vector in cone contrast space that represents chromaticity at the peak of the sinusoid (from the origin of the cone contrast space to peak contrast; equal to the square root of the sum of the squared cone contrasts). We used this contrast metric to make comparable the contrast-response curves obtained with stimuli modulated along different color directions. To study the effect of contrast, we used five different contrast levels for each stimulus type. These were as follows: Achromatic stimuli – 0.12, 0.24, 0.48, 0.72 and 0.96. L-M-cone stimuli – 0.015, 0.03, 0.06, 0.09 and 0.125. S-cone stimuli – 0.15, 0.325, 0.50, 0.675 and 0.85. 2.4. Spatiotemporal criteria To enhance the relative isolation of the different retino-cortical pathways, we used stimuli with distinct spatiotemporal criteria. Differences in temporal rates are not ideal, but they are also necessary due to the need of differentially recruiting populations with different tuning properties. Achromatic stimulus parameters: Spatial frequency 0.5 cycles per degree (cpd) and reversal rate of 10 rev/s. L-M-cone and S-cone stimulus parameters: Spatial frequency 2 cpd and reversal rate of 2 rev/s. 2.5. Experimental protocol

2.3. Manipulations of chromatic contrast We defined stimulus chromaticity using a three-dimensional cone contrast space in which each axis represents the activation of the L-, M- and S-cone types, normalized with respect to the white background (i.e. cone contrast) (Cole & Hine, 1992; Derrington, Krauskopf, & Lennie, 1984; MacLeod & Boynton, 1979). In this space, stimulus chromaticity is specified by a given vector direction. Krauskopf et al. (1982) defined three cardinal directions that exclusively stimulate each post-receptoral mechanism and no other and can be determined from the knowledge of the cone weights of each mechanism (Cole et al., 1993; Sankeralli & Mullen, 1996, 1997).

During EEG recording, stimulus presentation was divided into segments of around 10 min and subjects were allowed to rest in between segments as necessary. During each segment, only one type of stimulus was presented (achromatic, L-M-cone or S-cone stimuli) and stimulus contrast and size changed randomly every 3 s. This procedure minimized the likelihood of short-term adaptation effects. Furthermore, we did not observe any response drift for each stimulus condition across time. For each stimulus type there were five different contrasts, four different stimulus sizes and one blank condition (mean luminance background with fixation spot), making a total of 21 stimuli per retino-cortical pathway.

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

Participants were instructed to maintain fixation on a black spot at the centre of the screen while indicating through button presses if stimulus contrast increased or decreased. This behavioral task was aimed at maintaining the participant’s attention on the visual stimuli stable throughout the EEG recording session. 2.6. Data acquisition and analysis ERPs were recorded using a 64 channels Neuroscan system with scalp electrodes referenced to Cz and recalculated to linked ear references. Vertical and horizontal electrooculograms were also recorded in order to reject artifacts caused by blinking and eye movements. A trigger pulse was generated at the onset of each stimulus (at each phase reversal). Data analysis was performed with Scan4.3 (Neuroscan). For each subject, we chose to depict the analysis of the midline electrode with higher response amplitude (Pz, POz or Oz). We applied a bandpass filter with cutoff frequencies of 0.5 and 30 Hz, for the analysis of the L-M- and S-cone responses, and cutoff frequencies of 1 and 30 Hz for analysis of the achromatic response. All the filtering was performed using the 12 dB/octave and the Zero Phase Shift options available in Scan4.3. Zero Phase Shift filtering makes two ‘‘passes” through the filter, once in each direction and, therefore, it has no effect on the evoked potential component latencies. The recorded files were then cut into epochs. The epochs of the signal elicited by the achromatic stimuli were 200 ms long (1 full cycle at 5 Hz cycling rate) from beginning of each cycle. The epochs of the signals elicited by the L-M- and S-cone stimuli were 500 ms long, starting 50 ms before

A

8º

6º

4º 3. Results

1 V

0.125 0.090

Contrast

phase reversal and finishing 450 ms after. For L-M and S-cone stimulation, a baseline was set from 50 ms to the onset of the stimulus (0 ms). For achromatic stimulation, and given that we were measuring a steady state response, the baseline was determined as the average value of the entire sweep ranging from stimulus onset till 200 ms after, that is, our baseline represents the mean amplitude of one temporal cycle of the achromatic stimulus. Artifact rejections were conducted automatically on the basis of deflections with amplitude higher than 30 lV. The epochs of each stimulus type were averaged (around 100 for the achromatic response and around 150 for L-M- and S-cone responses). Note that, for calculation of the evoked potentials elicited by L-M- and S-cone stimuli, both phases of the stimulation cycle were averaged together. For the L-M- and S-cone responses, we calculated the amplitudes and latencies of the peaks by programming the Scan4.3 software to find automatically the maximums and minimums of the waves within time windows defined by inspection of the grand average VEPs. For the L-M-cone responses, we determined the amplitude and latency of the three positive peaks by imposing the following time windows: P1, the maximum between 50 and 110 ms; P2 the maximum between 120 and 180 ms; and P3, the maximum between 180 and 250 ms. For the S-cone responses, we measured the amplitude and latency of the three peaks, two maximums and one minimum: P1, the maximum between 50 and 150 ms; N1, the minimum between 100 and 200 ms; and P2, the maximum between 150 and 300 ms. The signal strength of the achromatic responses was not calculated based on peak amplitudes but rather on the mean amplitude of the rectified wave within a stimulus cycle. This was calculated also using the Scan4.3 software. Statistical analyses were performed in SPSS 15.0 for Windows.

Stimulus diameter 10º

0.060 0.030 0.015 0

200 400

0

ms

200 400

0

ms

B

200 400

0

200 400

ms

ms

Subj 3

Subj 4

Subjects Subj 1

Subj 2

0.125 0.090

Contrast

1285

0.060 0.030 0.015 0

200 400

ms

0

200 400

ms

0

200 400

ms

0

200 400

ms

Fig. 1. Visual evoked potentials (VEPs) obtained with L-M-cone modulations. Positive voltage is up. (A) Grand averages for five stimulus contrasts (labeled on the left) and four stimulus sizes (labeled on top). (B) Representative examples of individual VEPs from four participants, elicited by stimuli with 10° diameter and five different levels of contrast (labeled on the left).

Figs. 1a, 2a and 3a show the grand average of the VEPs of all participants elicited by L-M-cone, S-cone and achromatic stimulation, respectively. Figs. 1b, 2b and 3b show representative examples of individual VEPs from four participants elicited by the stimuli with largest diameter. The L-M- and S-cone stimuli elicited transient VEPs (Figs. 1 and 2, respectively), as expected from the temporal frequency of the stimuli (2 rev/s). The transient VEPs obtained with L-M- and S-cone stimulation show different characteristics. The grand average of the VEPs of all participants obtained with L-M-cone stimulation, suggested the presence of three positive components peaking at around 80, 130, and 210 ms after stimulus reversal. However, these three peaks were not always detected in individual VEPs, as it is apparent in the examples given in Fig. 1b. Statistical analyses of the amplitudes of these peaks revealed that only the amplitude of the first peak showed significant effects of stimulus contrast [F(1.6,21) = 7.4, p < 0.01] and size [F(2.7,35) = 9.2, p < 0.001]. The grand averages of the VEPs obtained with S-cone stimulation are as expected different showing a positive peak at around 100 ms, P1, followed by a negative peak at 140 ms, N1, and a positive peak at 200 ms, P2 (Fig. 2a). The three peaks can be distinguished in most individual VEPs, as shown in the examples (Fig. 2b). Statistical analyses of the peak amplitudes revealed significant effects of contrast and size for both P1 [effect of contrast: F(4,52) = 24.5, p < 0.001; effect of size: F(3,39) = 23.7, p < 0.001] and N1 [effect of contrast: F(1.68,38) = 16.3, p < 0.001; effect of size: F(3,39) = 8.58, p < 0.001], but not for P2. In subsequent analyses, we used peak to trough amplitude, P1–N1, as a measure of the response amplitude. The latency of the first VEP peak is consistently shorter for L-Mcone modulation when compared with the one elicited by S-cone

1286

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

A

A

Stimulus diameter 10º

8º

6º

4º

0.72

Contrast

Contrast

6º

0.96

0.675 0.500

0.48

0.325

0.24

0.150

0.12 0

200 400

0

ms

200 400

0

ms

B

200 400

0

ms

0

200 400

100

200 0

Subj 3

200 0

ms

B

Subj 2

100

ms

ms

Subjects Subj 1

100

200 0

100

200

ms

ms

Subj 3

Subj 4

Subjects Subj 1

Subj 4

Subj 2

0.96

0.850

0.72

Contrast

0.675

Contrast

8º 1 V

1 v

0.850

Stimulus diameter 10º

4º

0.500

0.48 0.24

0.325

0.12

0.150

0

200 400

ms

0

200 400

ms

0

200 400

ms

0

ms

Fig. 2. Visual evoked potentials (VEPs) obtained with S-cone modulations. Positive voltage is up. (A) Grand averages for five stimulus contrasts (labeled on the left) and four stimulus sizes (labeled on top). (B) Representative examples of individual VEPs from four participants, elicited by stimuli with 10° diameter and five different levels of contrast (labeled on the left).

modulation. We calculated the latency of this VEP component for all stimuli, except for the stimuli with lowest contrast given that the lowest contrast used for both L-M and S-cone stimulation did not elicit a detectable response (Figs. 1 and 2). Statistical analyses of the latency of the earliest peak showed a significant effect of stimulus contrast [L-M-cone stimulation: F(1.6,21) = 11.4, p < 0.01; S-cone stimulation: F(2.5,33) = 21.8, p < 0.001] but no effect of stimulus size [L-M-cone stimulation: F(3,39) = 1.46, p = 0.2; S-cone stimulation: F(1.6,20) = 3.66, p = 0.054]. Indeed, the latencies decreased steadily with stimulus contrast (Fig. 4). As it can be seen in Fig. 4, the average latency for any contrast is always longer for Scone stimulation when compared with L-M-cone stimulation. For statistical analyses, we took, for each participant, the average of the latency of the first peak elicited by S-cone or L-M-cone stimulation. The S-cone latencies were significantly longer than the L-Mcone latencies (paired t-test, p < 0.0001). In average, the difference in latency was 17 ms. A similar latency difference between the response to L-M-cone and S-cone signals has been observed in the macaque primary visual cortex (Cottaris & De Valois, 1998), in human VEP studies using pattern onset stimulation (Rabin, Switkes, Crognale, Schneck, & Adams, 1994; Robson & Kulikowski, 1998) and human psychophysics (Smithson & Mollon, 2004). In order to maximize the magnocellular contribution to the response of the achromatic channel, we used stimuli reversing at high frequency (10 rev/s) (Lee, 1996). The high temporal frequency of the achromatic stimuli elicited steady state VEPs with a cycling frequency equal to the reversal rate of the stimulus (double of the cycling frequency of the stimulus; Fig. 3). The shape of the positive part of the response suggests the presence of two largely overlap-

100

200 0

100

ms

200 400

200 0

ms

100

200 0

ms

100

200

ms

Fig. 3. Visual evoked potentials (VEPs) obtained with achromatic modulations. Positive voltage is up. (A) Grand averages for five stimulus contrasts (labeled on the left) and four stimulus sizes (labeled on top). (B) Representative examples of individual VEPs from four participants, elicited by stimuli with 10° diameter and five different levels of contrast (labeled on the left).

L-M-cone

115

Latency (ms)

0

S-cone

105

95

85

75 0

20

40

60

80

100

120

Contrast (% of maximum used) Fig. 4. Average VEP latencies as a function of contrast, for L-M and S-cone stimulation. Data are mean of all participants of the averages of responses to all stimulus sizes for each contrast. Error bars represent ±1 standard error. Contrasts are represented as percentage of maximum cone contrast used in each condition.

ping positive components. These two components could be detected in the majority of the individual VEPs, as can be seen in the examples given in Fig. 3b. As the two positive components overlap significantly, latency and amplitude of each of the peaks do not necessarily represent the latency and amplitude of the underlying components (Luck, 2004). Furthermore, for the analyses of the achromatic response, we did not separate the contribution of the different compo-

1287

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

nents. Instead, we measured the mean amplitude of the rectified wave over the period of one stimulus cycle; a measure that represents a steady state indicator of the strength of the signal evoked in the visual cortex (Luck, 2004). Repeated measures analyses, including the within subjects variables contrast and size, revealed that the mean amplitude of the achromatic response shows significant effects of contrast [F(1.24,16) = 17.0, p < 0.001], size [F(1.58,21) = 21,0, p < 0.001] and interaction across these dimensions [F(5.54,72) = 4.10, p < 0.01]. Fig. 5 shows the contrast-response functions of the human visual cortex elicited by the three different types of stimulation. For each stimulus type we used five different levels of contrast and four different stimulus sizes. For the L-M-cone response, we plotted the amplitude of the first positive peak (Fig. 5, top).

2.3

L-M-cone stimulation 1.9 4º 6º 8º 10º

1.5 1.1 0.7 0.3 0

0.05

0.1

6.0

Amplitude ( V)

5.0

S-cone stimulation

4º 6º 8º 10º

4.0 3.0 2.0 1.0 0.1

0.3

0.5

0.7

0.9

1.1

Achromatic stimulation 0.9 10º 8º 6º 4º

0.7

Repeated measures analyses, including the within subjects variables contrast and size, showed a significant effect of stimulus contrast and size, as reported above. Furthermore, post-hoc analysis of contrast-responses showed that the responses for the two highest contrasts were not significantly different (p = 0.7), suggesting response saturation. However, this lack of significance might, as an alternative explanation, be due to the intersubject variability that could be masking the linearity of the response. The amplitude of the S-cone response was taken as the peak-topeak amplitude, P1–N1. Repeated measures analyses, including the within subjects variables contrast and size, revealed that the peakto-peak amplitude showed a significant effect of contrast [F(1.6,20) = 42.1, p < 0.001] and size [F(1.4,18) = 31.7, p < 0.001] and a significant interaction between the effects of contrast and size [F(4.3,56) = 5.7, p < 0.001], reflecting the different slopes of the contrast-response functions for the different stimulus sizes (Fig. 5, middle). Post-hoc analysis of contrasts revealed a significant increase of the response for all stimulus contrasts (p < 0.01), suggesting that the response to S-cone modulation does not saturate within the levels of contrast used. Indeed, for all stimulus sizes, the responses are well fitted by linear functions (p < 0.01). The different slopes of the contrast-response curves obtained for the different stimulus sizes are what would be expected if a bigger stimulus was recruiting more neurons and each of these units increased its response linearly with contrast. The mean amplitude of the achromatic response showed saturating responses with contrast for all stimulus sizes used, as expected from previous reports of contrast-response functions elicited by achromatic stimulation. Indeed, post-hoc analysis of the contrastresponses showed a significant increase of the response with contrast (p < 0.05) except for the two highest contrasts (p = 0.2) where the responses were not significantly different, in agreement with response saturation. The achromatic response showed a significant effect of interaction between the variables contrast and size, probably reflecting, the increase in slope with stimulus size observed and that had been described before for achromatic stimulation (Wright & Johnston, 1982). However, the shape of the contrast-response functions did not change with stimulus size. The increase in amplitude with size probably follows the increase in number of neurons activated by the stimulus in accordance with the retinotopic organization of the early visual system. As neuronal density is constant within the visual cortex (Rockel, Hiorns, & Powell, 1980), the number of cortical neurons activated is proportional to the area of the primary visual cortex activated by the stimuli. Polimeni, Balasubramanian, and Schwartz (2006) proposed a mathematical representation of the area of V1 activated by achromatic circular stimuli centered in the fovea. Using their model, we calculated how this cortical area varies as a function of stimulus diameter. According to the cortical magnification factor, the area of V1 activated increases logarithmically with stimulus diameter (Fig. 6). In Fig. 7, we plotted the amplitude of the visual evoked potentials versus the natural logarithm of the radius of the stimuli. The size response functions obtained with S-cone and achromatic stimula-

0.3 0.1

0.3

0.5

0.7

0.9

Cone contrast Fig. 5. Average VEP amplitude as a function of cone contrast, for four different stimulus sizes (approximate stimulus diameters shown in legend next to each graphic). The top graphic shows the responses elicited by L-M-cone stimulation; the middle graphic the responses elicited by S-cone stimulation and the bottom graphic the responses elicited by achromatic stimulation. Error bars represent standard errors of the mean. Both the L-M-cone and achromatic stimuli elicited saturating contrast-response functions while the S-cone stimulation evoked linear contrastresponse functions.

Area (mm2)

0.5 1600 1200

y = 404.62Ln(x) + 145.72 R2 = 0.9994

800 400 0 0

10

20

30

40

Radius of stimulus (º) Fig. 6. Area of primary visual cortex activated by an achromatic circular stimulus, presented in the centre of the visual field, as a function of stimulus radius. Data points were determined using the model proposed by Polimeni et al. (2006). Dashed line is best logarithmic fitting. Equation and R-square of logarithmic fit are shown.

1288

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

2

L-M-cone stimulation 0.125 0.09 0.06 0.03 0.015

1

0 0.6

0.8

1

1.2

1.4

1.6

6

S-cone stimulation Amplitude ( V)

5 0.15 0.325 0.5 0.675 0.85

4 3 2 1 0.6

1

1.1

1.6

4. Discussion

Achromatic stimulation

0.8

0.96 0.72 0.48 0.24 0.12

0.6

0.4

0.2 0.6

were 0.98 ± 0.14 lV (mean ± standard error) higher than the responses elicited by the smallest stimuli; while the increase in L-Mcone responses with stimulus size was 0.49 ± 0.09 lV. This difference was statistically significant (paired t-test, p < 0.01), suggesting that the effect of size on cortical response obeys different rules for LM-cone mechanism and S-cone mechanism. It has been suggested that chromatic pattern-reversal stimulation elicits VEPs of low signal to noise ratio and that onset-offset stimulation is more appropriate to probe the chromatic channels (Rabin et al., 1994; Robson, Holder, Moreland, & Kulikowski, 2006). Rabin et al. (1994) suggested that the difference between the responses to these two types of stimulation occurs due to different levels of contrast adaptation. In order to verify if, during our L-M-cone stimulation protocol, cortical responses were diminishing due to contrast habituation, we measured evoked potentials elicited during the first 5 min of stimulation (early responses) and compared these with the potentials evoked during the second 5 min of stimulation (late responses). We expected to detect a reduction in the amplitude of the late evoked potentials if adaptation was occurring. However, within subjects repeated measures analysis revealed no significant difference between early and late evoked potentials [F(1,13) = 1.0, p = 0.3]. As expected, the amplitudes of evoked potentials calculated in this manner showed significant effects of stimulus contrast [F(4,52) = 3.8, p < 0.01] and size [F(3,39) = 4.7, p < 0.01]. These results suggest that our stimulation protocol, with stimulus contrast and/or size changing randomly every 3 s, does not induce significant contrast adaptation.

0.8

1

1.2

1.4

1.6

Ln radius Fig. 7. Average VEP amplitude as a function of stimulus size, for each contrast tested, obtained with L-M-cone stimulation (top), S-cone stimulation (middle) and achromatic stimulation (bottom). Each line joins data obtained with the same stimulus contrast but different stimulus sizes. The contrasts used are shown in the legend on the right of the graphs. Error bars represent standard errors of the mean.

tion were well described by linear functions of the logarithm of the radius, except for the lowest stimulus contrast where the responses did not increase significantly with size (achromatic linear size response functions for increasing contrasts: r = 0.852, r = 0.995, r = 0.973, r = 0.978 and r = 0.998; S-cone: r = 0.676, r = 0.996, r = 0.992, r = 0.985 and r = 0.995). For all contrasts tested, the L-Mcone modulation also elicited a response that showed a significant effect of stimulus size [F(2.7,35) = 9.2, p < 0.001]. However, the linear fit of the size response functions elicited by L-M-cone modulation did not explain as much variance as the S-cone and achromatic responses (L-M-cone linear size response functions for increasing contrasts: r = 0.886, r = 0.962, r = 0.872, r = 0.968 and r = 0.835). The increase in response amplitude with size also seems to be smaller for the L-M-cone stimulation than for S-cone stimulation. To make the responses comparable, we measured the increase in peak amplitude for the first positive peak of the L-M- and S-cone stimulation, taking the average of the responses for all contrasts elicited by each stimulus size. The S-cone responses elicited by the biggest stimuli

One of the main findings of our paper was the identified correlation between previously characterized psychophysical mechanisms and population responses in the visual cortex. Such a fundamental correlation could be established based on physiological properties of the chromatic and achromatic visual channels. Contrast-response functions were distinct for each of the pathways and consistent with what is known from single cell animal studies and population responses as inferred from human psychophysics and fMRI. Most importantly, our data suggests a parallel between the achromatic and S-cone pathways, in spatial summation, that might indicate a similar role for the S-cone pathway in peripheral spatial vision. 4.1. Area summation As expected, both the chromatic and achromatic post-receptoral mechanisms showed significant effects of stimulus size. Interestingly, there were differences between the area summation observed in the L-M-cone response and the other two mechanisms. The responses to luminance and S-cone stimulation were proportional to the area of visual cortex activated suggesting that their responses did not change across eccentricities and were proportional to the number of neurons stimulated. This observation suggests that the retinotopic distribution of neurons responding to S-cone contrast in visual cortex is similar to the distribution of neurons that respond to luminance contrast, decreasing with the logarithm of the distance to the fovea. Accordingly, fMRI studies have shown that the spatial response of V1 to blue-yellow stimulation is similar to the response to achromatic stimulation, in that both response types vary little with eccentricity (Mullen et al., 2007; Vanni et al., 2006). In contrast, the average L-M-cone response did not correlate with the area of cortex activated and the increase in the response elicited by L-M-cone stimulation with stimulus size was significantly smaller than for S-cone stimulation. This smaller effect of stimulus size is consistent with the foveal bias of the L-Mcone mechanism suggested by psychophysical, functional and ana-

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

tomical data (Baseler & Sutter, 1997; Mullen & Kingdom, 2002; Mullen, Sakurai, & Chu, 2005; Mullen et al., 2007; Sakurai & Mullen, 2006; Silva et al., 2008; Vanni et al., 2006; Yamada, Silveira, Perry, & Franco, 2001). Furthermore, the L-M-cone response did not show an interaction between the effects of size and contrast. In contrast, there was an effect of stimulus contrast on the size response functions elicited by S-cone and luminance stimulation. The lack of interaction between effects of contrast and size in the response to L-M stimulation suggests a different interaction between area summation and contrast response probably due to different properties in the L-M-cone contrast response in the central and peripheral visual fields. In other words, we provide additional evidence suggesting that, within our range of eccentricities, the spatial processing of S-cone and achromatic mechanisms are closer as compared with the L-M-cone channel. However, it is possible that the similarities between the S-cone and luminance channels emerge due to activation of the luminance channel by the S-cone stimuli. Indeed, VEP studies have indicated luminance intrusion in large S-cone isolating stimuli (Kulikowski, Robson, & McKeefry, 1996; Robson et al., 2006). Yet, we believe that this luminance intrusion is minimal (Switkes, Crognale, Rabin, Schneck, & Adams, 1996). Indeed, Mullen et al. (2007) used large, 12° diameter, S-cone isolating stimuli in an fMRI study. Their results strongly suggest that the cortical response elicited by their S-cone stimuli comes largely from the S-cone channel and not the achromatic channel. For instance, S-cone stimulation did not activate MT+, a dorsal stream area that receives strong input from the luminance channel, even at low luminance contrasts, but responds poorly to chromatic stimuli (Liu & Wandell, 2005; Tootell et al., 1995). This evidence indicates that, even if there is some luminance intrusion in our S-cone isolating stimuli, the majority of the cortical response reflects activation of the S-cone channel. 4.2. Latency differences The responses to the two types of chromatic stimulation peaked at different latencies after stimulus reversal, with the responses to L-M-cone modulation being faster than the S-cone responses. We compared latencies obtained with higher S-cone contrasts with latencies obtained with lower L-M-cone contrasts. The two ranges of contrasts were different because the S-cone pathway is less sensitive than the L-M pathway and L-M-cone isoluminance can only be obtained at low cone contrasts. However, because the S-cone responses showed higher latencies for higher cone contrasts and the latencies decreased significantly with contrast for both channels, then if we had used lower S-cone contrasts equivalent to the low L-M-cone contrasts used, the latency difference would be even higher and, therefore, still significant. Furthermore, the differences in latencies observed are consistent with previous observation in the macaque primary visual cortex (Cottaris & De Valois, 1998), in human VEP studies (Rabin et al., 1994; Robson & Kulikowski, 1998) and human psychophysics (Smithson & Mollon, 2004; Wade, 2009), confirming that, in this study, we are probing the two channels separately. Cortical signal amplification has been proposed as an explanation for the slow cortical response to S-cone stimulation (Cottaris & De Valois, 1998). Cortical amplification of the S-cone response was supported by a recent study that showed that the response to S-cone stimulation is relatively weak in sub-cortical regions but strong in the cortex (Mullen, Dumoulin, & Hess, 2008). In our study, we also observed a strong S-cone response consistent with signal amplification. 4.3. Contrast responses The saturating contrast-response functions elicited by L-M-cone stimulation were unexpected. L-M-cone stimulation activates

1289

mostly the parvocellular pathway and the response of LGN parvocellular neurons is approximately proportional to stimulus contrast (Kaplan & Shapley, 1982; Solomon & Lennie, 2005). However, the response of cortical neurons to chromatic stimulation is more variable. In fact, for visual stimulation that modulates L-M-cone contrast, a significant proportion of V1 neurons show contrast-response functions that saturate at medium contrast levels (Solomon & Lennie, 2005). Consistent with our findings, other VEP and fMRI studies revealed saturating contrast-response functions elicited by isoluminant red-green gratings (Alexander, Rajagopalan, Seiple, Zemon, & Fishman, 2005; Porciatti, Bonanni, Fiorentini, & Guerrini, 2000; Regan, 1973; Vanni et al., 2006). On the contrary, other VEP studies have described linear red-green chromatic contrast-response functions (Greenstein, Seliger, Zemon, & Ritch, 1998; Porciatti & Sartucci, 1999; Souza et al., 2008). Differences in stimulation paradigms, stimulus chromaticities, range of contrasts used or methods for amplitude calculation might be the cause for this difference. Furthermore, it is possible that the intersubject variability of the responses present in our data has masked a putatively more linear contrast response function. In addition, temporal contrast adaptation mechanisms could be a potential alternative explanation for the saturating contrast-response functions. However, we did not find any evidence for such effect in our data. Unlike luminance and L-M-cone contrasts, S-cone contrast elicited cortical responses proportional to stimulus contrast with no evidence of saturation. This result is consistent with the linear response of koniocellular neurons sensitive to S-cone modulation in the LGN (Tailby, Solomon, & Lennie, 2008). Moreover, in the V1 of the macaque, the majority of cells that responded strongly to S-cone modulation showed linear contrast-response functions (Solomon & Lennie, 2005). In fact, in agreement with our study, modulation of Scone excitation elicited the most linear responses in V1 (Solomon & Lennie, 2005). Furthermore, the absence of saturation in our S-cone contrast responses is in accordance with a contrast discrimination study that showed no masking within our levels of S-cone contrast (Wade, 2009). Previous studies have, however, reported responses to S-cone stimulation that deviate from linearity (Rabin et al., 1994; Vanni et al., 2006). It is possible that some of the inconsistencies found are due to differences in stimulus properties. Rabin et al. (1994) measured VEPs of onset visual presentation, while, in our study and the single cell recordings mentioned, the visual stimuli were phase reversing gratings. Vanni et al. (2006) used multifocal fMRI to study the response of human V1 to chromatic modulation and found that cortical population responses to S-cone modulation show saturation that, nevertheless, deviates less from linearity than the one observed for L-M-cone modulation. Therefore, most likely, all three channels will show saturation of the response although in distinct amounts when compared across similar stimuli. In any case, some level of saturation should be present which is consistent with contrast discrimination studies revealing the effect of masking for all three channels, suggestive of compression of the respective contrast-response functions (Campbell & Kulikowski, 1966; Chen, Foley, & Brainard, 2000; Losada & Mullen, 1994). Onset presentation has been shown to elicit stronger chromatic VEP responses, as opposed to pattern-reversal stimulation (Rabin et al., 1994). Furthermore, Robson et al. (2006) suggested that evoked potentials elicited by pattern-reversal chromatic stimulation reflected the activity of luminance and not chromatic channels. However, fMRI studies using pattern-reversal stimulation showed distinct patterns of brain activation for achromatic and chromatic patterns (Liu & Wandell, 2005; Vanni et al., 2006), suggesting that, contrary to what was proposed, pattern-reversal chromatic stimulation activates the chromatic channels. In addition, previous VEP studies have shown that chromatic pattern-reversal stimulation elicits significant evoked potentials (Alexander et al., 2005; Baseler & Sutter, 1997; Gomes et al., 2008; Kulikowski

1290

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291

et al., 1996; Porciatti, Di Bartolo, Nardi, & Fiorentini, 1997; Rudvin & Valberg, 2005; Souza et al., 2008). Therefore, we believe that although onset pattern presentation evokes chromatic responses with larger signal to noise ratio, pattern-reversal mode of stimulation is also appropriate to probe chromatic channels. We conclude that psychophysical chromatic and achromatic detection mechanisms can be tightly linked with population responses in human visual cortex in a way that reflects the response properties of ensembles of visual neurons. Acknowledgments This work was supported by the University of Coimbra (Grant reference: III/14/2008), Portuguese Foundation for Science and Technology (Grant references: SFRH/BPD/34392/2006, PIC/IC/ 83155/2007 and PTDC_SAU_NEU_68483_2006) and National Brain Imaging Network of Portugal (Grant reference: REDE/1519/RNIFC/ 2006). References Albrecht, D. G., & Hamilton, D. B. (1982). Striate cortex of monkey and cat: Contrast response function. Journal of Neurophysiology, 48(1), 217–237. Alexander, K. R., Rajagopalan, A. S., Seiple, W., Zemon, V. M., & Fishman, G. A. (2005). Contrast response properties of magnocellular and parvocellular pathways in retinitis pigmentosa assessed by the visual evoked potential. Investigative Ophthalmology and Visual Science, 46(8), 2967–2973. Baseler, H. A., & Sutter, E. E. (1997). M and P components of the VEP and their visual field distribution. Vision Research, 37(6), 675–690. Boon, M. Y., Suttle, C. M., & Henry, B. (2005). Estimating chromatic contrast thresholds from the transient visual evoked potential. Vision Research, 45(18), 2367–2383. Boynton, G. M., Demb, J. B., Glover, G. H., & Heeger, D. J. (1999). Neuronal basis of contrast discrimination. Vision Research, 39(2), 257–269. Brainard, D. H., Pelli, D. G., & Robson, T. (2002). Display characterization. Encylopedia of imaging science and technology. Wiley (pp. 172–188). Campbell, F. W., & Kulikowski, J. J. (1966). Orientational selectivity of the human visual system. Journal of Physiology, 187(2), 437–445. Cavanagh, P., MacLeod, D. I., & Anstis, S. M. (1987). Equiluminance: Spatial and temporal factors and the contribution of blue-sensitive cones. Journal of the Optical Society of America A, Optics and Image Science, 4(8), 1428–1438. Chen, C., Foley, J. M., & Brainard, D. H. (2000). Detection of chromoluminance patterns on chromoluminance pedestals. I: Threshold measurements. Vision Research, 40(7), 773–788. Cole, G. R., & Hine, T. (1992). Computations of cone contrasts for color vision research. Behaviour Research Methods, Instruments and Computers, 24(1), 22–27. Cole, G. R., Hine, T., & McIlhagga, W. (1993). Detection mechanisms in L-, M-, and Scone contrast space. Journal of the Optical Society of America A, Optics and Image Science, 10(1), 38–51. Cottaris, N. P., & De Valois, R. L. (1998). Temporal dynamics of chromatic tuning in macaque primary visual cortex. Nature, 395(6705), 896–900. Derrington, A. M., Krauskopf, J., & Lennie, P. (1984). Chromatic mechanisms in lateral geniculate nucleus of macaque. Journal of Physiology, 357, 241–265. Foley, J. M., Varadharajan, S., Koh, C. C., & Farias, M. C. (2007). Detection of Gabor patterns of different sizes, shapes, phases and eccentricities. Vision Research, 47(1), 85–107. Gomes, B. D., Souza, G. S., Lima, M. G., Rodrigues, A. R., Saito, C. A., da Silva Filho, M., et al. (2008). Color discrimination ellipses of trichromats measured with transient and steady state visual evoked potentials. Visual Neuroscience, 25(3), 333–339. Greenstein, V. C., Seliger, S., Zemon, V., & Ritch, R. (1998). Visual evoked potential assessment of the effects of glaucoma on visual subsystems. Vision Research, 38(12), 1901–1911. Hendry, S. H., & Reid, R. C. (2000). The koniocellular pathway in primate vision. Annual Review of Neuroscience, 23, 127–153. Kaplan, E., & Shapley, R. M. (1982). X and Y cells in the lateral geniculate nucleus of macaque monkeys. Journal of Physiology, 330, 125–143. Krauskopf, J., Williams, D. R., & Heeley, D. W. (1982). Cardinal directions of color space. Vision Research, 22(9), 1123–1131. Kulikowski, J. J., Robson, A. G., & McKeefry, D. J. (1996). Specificity and selectivity of chromatic visual evoked potentials. Vision Research, 36(21), 3397–3401. Lee, B. B. (1996). Receptive field structure in the primate retina. Vision Research, 36(5), 631–644. Liu, J., & Wandell, B. A. (2005). Specializations for chromatic and temporal signals in human visual cortex. Journal of Neuroscience, 25(13), 3459–3468. Losada, M. A., & Mullen, K. T. (1994). The spatial tuning of chromatic mechanisms identified by simultaneous masking. Vision Research, 34(3), 331–341. Luck, S. J. (2004). Ten simple rules for designing ERP experiments. In T. C. Handy (Ed.), Event-related potentials: A methods handbook (pp. 17–32). Cambridge, MA: MIT Press.

MacLeod, D. I., & Boynton, R. M. (1979). Chromaticity diagram showing cone excitation by stimuli of equal luminance. Journal of the Optical Society of America, 69(8), 1183–1186. Meese, T. S., Hess, R. F., & Williams, C. B. (2005). Size matters, but not for everyone: Individual differences for contrast discrimination. Journal of Vision, 5(11), 928–947. Merigan, W. H., & Maunsell, J. H. (1993). How parallel are the primate visual pathways? Annual Review of Neuroscience, 16, 369–402. Mullen, K. T., Dumoulin, S. O., & Hess, R. F. (2008). Color responses of the human lateral geniculate nucleus: [corrected] Selective amplification of Scone signals between the lateral geniculate nucleno and primary visual cortex measured with high-field fMRI. European Journal of Neuroscience, 28(9), 1911–1923. Mullen, K. T., Dumoulin, S. O., McMahon, K. L., de Zubicaray, G. I., & Hess, R. F. (2007). Selectivity of human retinotopic visual cortex to S-cone-opponent, L/Mcone-opponent and achromatic stimulation. European Journal of Neuroscience, 25(2), 491–502. Mullen, K. T., & Kingdom, F. A. (2002). Differential distributions of red-green and blue-yellow cone opponency across the visual field. Visual Neuroscience, 19(1), 109–118. Mullen, K. T., Sakurai, M., & Chu, W. (2005). Does L/M cone opponency disappear in human periphery? Perception, 34(8), 951–959. Page, J. W., & Crognale, M. A. (2005). Differential aging of chromatic and achromatic visual pathways: Behavior and electrophysiology. Vision Research, 45(11), 1481–1489. Polimeni, J. R., Balasubramanian, M., & Schwartz, E. L. (2006). Multi-area visuotopic map complexes in macaque striate and extra-striate cortex. Vision Research, 46(20), 3336–3359. Porciatti, V., Bonanni, P., Fiorentini, A., & Guerrini, R. (2000). Lack of cortical contrast gain control in human photosensitive epilepsy. Nature Neuroscience, 3(3), 259–263. Porciatti, V., Di Bartolo, E., Nardi, N., & Fiorentini, A. (1997). Responses to chromatic and luminance contrast in glaucoma: A psychophysical and electrophysiological study. Vision Research, 37(14), 1975–1987. Porciatti, V., & Sartucci, F. (1999). Normative data for onset VEPs to red-green and blue-yellow chromatic contrast. Clinical Neurophysiology, 110(4), 772–781. Rabin, J., Switkes, E., Crognale, M., Schneck, M. E., & Adams, A. J. (1994). Visual evoked potentials in three-dimensional color space: Correlates of spatiochromatic processing. Vision Research, 34(20), 2657–2671. Regan, D. (1973). Evoked potentials specific to spatial patterns of luminance and colour. Vision Research, 13(12), 2381–2402. Ress, D., & Heeger, D. J. (2003). Neuronal correlates of perception in early visual cortex. Nature Neuroscience, 6(4), 414–420. Robson, A. G., Holder, G. E., Moreland, J. D., & Kulikowski, J. J. (2006). Chromatic VEP assessment of human macular pigment: Comparison with minimum motion and minimum flicker profiles. Visual Neuroscience, 23(2), 275–283. Robson, A. G., & Kulikowski, J. J. (1998). Objective specification of tritanopic confusion lines using visual evoked potentials. Vision Research, 38(21), 3499–3503. Rockel, A. J., Hiorns, R. W., & Powell, T. P. (1980). The basic uniformity in structure of the neocortex. Brain, 103(2), 221–244. Rudvin, I., & Valberg, A. (2005). Visual evoked potentials for red-green gratings reversing at different temporal frequencies: Asymmetries with respect to isoluminance. Visual Neuroscience, 22(6), 735–747. Sakurai, M., & Mullen, K. T. (2006). Cone weights for the two cone-opponent systems in peripheral vision and asymmetries of cone contrast sensitivity. Vision Research, 46(26), 4346–4354. Sankeralli, M. J., & Mullen, K. T. (1996). Estimation of the L-, M- and S-cone weights of the post-receptoral detection mechanisms. Journal of the Optical Society of America A, Optics and Image Science, 13, 906–915. Sankeralli, M. J., & Mullen, K. T. (1997). Postreceptoral chromatic detection mechanisms revealed by noise masking in three-dimensional cone contrast space. Journal of the Optical Society of America A, Optics and Image Science, 14(10), 2633–2646. Silva, M. F., Maia-Lopes, S., Mateus, C., Guerreiro, M., Sampaio, J., Faria, P., et al. (2008). Retinal and cortical patterns of spatial anisotropy in contrast sensitivity tasks. Vision Research, 48(1), 127–135. Sincich, L. C., Park, K. F., Wohlgemuth, M. J., & Horton, J. C. (2004). Bypassing V1: A direct geniculate input to area MT. Nature Neuroscience, 7(10), 1123–1128. Smithson, H. E., & Mollon, J. D. (2004). Is the S-opponent chromatic sub-system sluggish? Vision Research, 44(25), 2919–2929. Solomon, S. G., Lee, B. B., & Sun, H. (2006). Suppressive surrounds and contrast gain in magnocellular-pathway retinal ganglion cells of macaque. Journal of Neuroscience, 26(34), 8715–8726. Solomon, S. G., & Lennie, P. (2005). Chromatic gain controls in visual cortical neurons. Journal of Neuroscience, 25(19), 4779–4792. Souza, G. S., Gomes, B. D., Lacerda, E. M., Saito, C. A., da Silva Filho, M., & Silveira, L. C. (2008). Amplitude of the transient visual evoked potential (tVEP) as a function of achromatic and chromatic contrast: Contribution of different visual pathways. Visual Neuroscience, 25(3), 317–325. Stockman, A., & Sharpe, L. T. (2000). The spectral sensitivities of the middle- and long-wavelength-sensitive cones derived from measurements in observers of known genotype. Vision Research, 40(13), 1711–1737. Switkes, E., Crognale, M., Rabin, J., Schneck, M. E., & Adams, A. J. (1996). Reply to ‘‘specificity and selectivity of chromatic visual evoked potentials”. Vision Research, 36(21), 3403–3405.

M.J. Ribeiro, M. Castelo-Branco / Vision Research 50 (2010) 1283–1291 Tailby, C., Solomon, S. G., & Lennie, P. (2008). Functional asymmetries in visual pathways carrying S-cone signals in macaque. Journal of Neuroscience, 28(15), 4078–4087. Tootell, R. B., Reppas, J. B., Kwong, K. K., Malach, R., Born, R. T., Brady, T. J., et al. (1995). Functional analysis of human MT and related visual cortical areas using magnetic resonance imaging. Journal of Neuroscience, 15(4), 3215–3230. Vanni, S., Henriksson, L., Viikari, M., & James, A. C. (2006). Retinotopic distribution of chromatic responses in human primary visual cortex. European Journal of Neuroscience, 24(6), 1821–1831. Wade, A. R. (2009). Long-range suppressive interactions between S-cone and luminance channels. Vision Research, 49(12), 1554–1562.

1291

Wandell, B. A., Poirson, A. B., Newsome, W. T., Baseler, H. A., Boynton, G. M., Huk, A., et al. (1999). Color signals in human motion-selective cortex. Neuron, 24(4), 901–909. Wright, M. J., & Johnston, A. (1982). The effects of contrast and length of gratings on the visual evoked potential. Vision Research, 22(11), 1389–1399. Wright, M. J., & Johnston, A. (1983). Spatiotemporal contrast sensitivity and visual field locus. Vision Research, 23(10), 983–989. Yamada, E. S., Silveira, L. C., Perry, V. H., & Franco, E. C. (2001). M and P retinal ganglion cells of the owl monkey: Morphology, size and photoreceptor convergence. Vision Research, 41(2), 119–131.