Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Neuron. Author manuscript; available in PMC 2007 Sep 20.
Published in final edited form as:
PMCID: PMC1988694

Compression and reflection of visually evoked cortical waves


Neuronal interactions between primary and secondary visual cortical areas are important for visual processing, but the spatiotemporal patterns of the interaction are not well understood. We used voltage-sensitive dye imaging to visualize neuronal activity in rat visual cortex and found novel visually evoked waves propagating from V1 to other visual areas. A primary wave originated in the monocular area of V1 and was “compressed” when propagating to V2. A reflected wave initiated after compression and propagated backward into V1. The compression occurred at the V1/V2 border, and local GABAA inhibition is important for the compression. The compression/reflection pattern provides a two-phase modulation: V1 is first depolarized by the primary wave and then V1 and V2 are simultaneously depolarized by the reflected and primary waves, respectively. The compression/reflection pattern only occurred for evoked but not for spontaneous waves, suggesting that it is organized by an internal mechanism associated with visual processing.

Keywords: voltage-sensitive dye, propagating waves, visual cortex, wave compression, wave reflection


During visual processing, extensive interactions occur both within the primary visual cortex (V1) and between visual areas via feedforward and feedback projections (Rockland and Pandya, 1981; Kennedy and Bullier, 1985; Livingstone and Hubel, 1987, 1988; Angelucci et al., 2002; Sincich and Horton, 2002a, 2002b, 2003; Shmuel et al., 2005). Such intra- and inter-areal interactions may follow a stereotypical spatial pattern and temporal sequence between the visual areas, and may manifest as propagation of excitation waves at the population level. In invertebrates and lower vertebrates, propagating waves have been suggested to participate in visual and olfactory processing (Delaney et al., 1994; Prechtl et al., 1997, 2000; Senseman and Robbins, 1999; Lam et al., 2000, 2003). In mammals, propagating waves have also been observed in somatosensory cortex and olfactory bulb (Freeman and Barrie, 2000; Derdikman et al. 2003; Petersen et al., 2003a, b; Civillico and Contreras, 2006; Ferezou et al., 2006). In motor cortex, waves were suggested to mediate information transfer during movement preparation and execution (Rubino et al., 2006). However, in mammalian visual cortex, while waves were reported in a few studies (Arieli et al., 1995; Roland et al., 2006), the spatiotemporal patterns of evoked waves have not been carefully examined. Since propagating waves determine when and where population depolarization will occur in the cortical network, they may play critical roles in cortical processing (Ermentrout and Kleinfeld, 2001; Rubino et al., 2006). Thus, characterizing the initiation and spatiotemporal patterns of the evoked waves in visual areas is important for understanding the population mechanisms of visual processing.

Voltage-sensitive dye imaging provides a useful tool for visualizing the spatiotemporal patterns of cortical activity. With the improvement of blue dyes (Shoham et al., 1999), sensory evoked activity can be observed from mammalian cortex in vivo with high signal-to-noise ratio (Derdikman et al. 2003; Petersen et al. 2003a, b; Grinvald and Hildesheim, 2004; Ferezou et al., 2006; Chen et al., 2006). In this report, we used voltage-sensitive dye imaging to examine visually evoked activity in rat visual cortical areas. Our imaging device offers 17-19 bit dynamic range, allowing us to examine wave dynamics in detail in single trials (Lippert et al., 2007). We found that visual stimulus initiated a propagating wave in V1, which was compressed when propagating to V2. A reflected wave was subsequently initiated and propagated back into V1. Further study showed that the compression occurred at the V1/V2 border. Similar compression/reflection patterns were also observed at the border between mediomedial V2 (V2MM) and retrosplenial dysgranular (RSD) areas. These compression/reflection patterns occur only in evoked but not in spontaneous waves, suggesting that the compression and reflection are governed by mechanisms specific for processing visual inputs.


Evoked waves: Compression and reflection

Voltage-sensitive dye signals were measured with a photodiode array from V1 and V2 areas of anesthetized rat (Figure 1A). The visual stimulus was a drifting grating (0.05 cycle/degree, 50w × 38h degrees of viewing angle) presented to the contralateral eye. The drift of the grating (3 cycles/sec) reliably evoked a propagating wave in the visual cortex. The evoked wave initiated with a latency of ∼100 ms (99.8 ± 18.2 ms, mean ± SD, n = 115 trials) after the onset of the drifting, and the activity was seen in all optical detectors with a small time difference between each detector (Figure 1B, traces 1- 4). The signal on each individual detector was converted to pseudo-color according to a linear color scale. The pseudo-color images showed that the evoked wave initiated in the monocular area of V1 (V1M) and propagated in both directions to the V1 binocular area (V1B) and to V2. This evoked wave, referred to as the primary wave, was “compressed” in its spatial dimension into a thin band in the middle of the propagating path (Figure 1C). A reflected wave initiated after compression and propagated backward to V1 (Figure 1C). The primary and reflected waves can be identified in the signal traces of individual detectors as double peaks (Figure 1B). Supplemental movie 1 presents another example showing the spatiotemporal sequence of the compression/reflection.

Figure 1
Wave compression and reflection

This compression/reflection pattern was reliably observed in different recording trials. Figure 2A shows wave patterns from the same animal with identical stimuli (inter-trial interval of ∼ 200 sec). In this animal, the compression bands reached the narrowest width (Figure 2A column c) at 72.7 ± 7.2 ms (mean ± SD, n = 9) after the onset of the primary wave. The compression band then became wider again due to the wave propagating into V2 and the back propagation of the reflected wave (Figure 2A column d). The location and the shape of the compression band were similar from trial to trial. Supplemental movie 2 provides an example from another animal, in which three trials show almost identical location and temporal sequence for the wave compression. We have examined the visually evoked waves in 36 animals. A similar primary wave, compression band, and reflected wave were observed in all animals. In Figure 2B, representative trials from seven animals all show similar spatiotemporal patterns. In different animals, the locations of the compression and the shape of the compression band varied slightly, probably reflecting individual variability in the neuroanatomy of the visual areas.

Figure 2
Wave compression occurs robustly

The similar compression/reflection pattern was observed under visual stimuli with various parameters, including alternation of orientation (0, 90, 180, 270°), drifting velocity (30 - 200 deg/sec), spatial frequency (0.025 - 0.3 cycles/degree), stimulus position (supplemental Figure S3), contrast (>0.5, supplemental Figure S4) and size (>10°, supplemental Figure S4), while the probability for initiating the primary wave, the initiation site and the shape of the compression band can be altered by varying stimulus parameters. Changes in stimulus position altered the location of the primary wave initiation site, following the retinotopic map in the V1M (supplemental Figure S2, S3A). The shape of the compression band also varied when the location of the initiation site changed (red and blue lines in supplemental Figure S3A). The probability of evoking the wave decreased when either stimulus size or contrast is reduced, with a threshold of 6 - 10° and 0.2 - 0.5 respectively (supplemental Figure S4). However, once the primary wave was initiated by supra-threshold stimulation, the same pattern of compression/reflection occurred. This was true even when the stimulus was presented at two positions with large difference in the visual field (supplemental Figure S3B, top and middle row). Thus, the wave compression/reflection is the rule rather than the exception.

Compression at the border between visual areas?

The reliability of the wave compression suggests it may be related to the neuroanatomical structure of the cortex, especially the border between V1 and V2. To test this idea, we used corpus callosum fiber bundles to identify the V1/V2 border. In rats, these bundles are abundant near the V1/V2 border (Olavarria and Hiroi, 2003), so when electrical shocks were applied to the visual cortex contralateral to the imaging side, action potentials may reach the imaging side by the callosal fibers and be visualized with voltage-sensitive dye imaging. Indeed, electrical stimulation in the contralateral V1M area with a moderate intensity evoked a localized activity on the imaging side (Figure 3B). The activity loci on the imaged side were fixed when the stimulation site was fixed, and the post stimulus latency of the activity was short and fixed (22.4 ± 2.6 ms, mean ± SD, n = 12 trials from 3 animals), indicating that electrical shock evoked the activity on the imaging side via callosal fiber bundles. In the same animal, visually evoked waves (Figure 3A) compressed adjacent to the activity evoked by contralateral electrical shocks (Figure 3C,D), suggesting that the compression occurred near the V1/V2 border.

Figure 3
Compression band and the corpus callosum fibers

We next examined if compression occurs at borders between other visual areas. Indeed, a second compression often occurred (observed in 11 out of 36 animals) along the propagating path (Figure 4A). The location of the second compression (Figure 4C) was more medial to that of the first compression (Figure 4B), probably at the border between V2MM and RSD areas. Between the two compressions there was a narrow gap, correlated well with the V2MM area (Figure 4C). While the onset time of the second compression was more variable from trial to trial, the location of the second compression was fixed. Multiple compressions suggest that wave compression is associated with the border between visual cortical areas.

Figure 4
Multiple compressions along the propagating path

Mechanisms of wave compression

The compression of the primary wave started as an abrupt slowing of the wave leading edge. As shown in Figure 5, the primary wave was initiated by the visual stimulus and quickly expanded into the entire V1 area (Figure 5A, first 2 images) at a propagation velocity of 50 -70 mm/s. When reaching the V1-V2 border, the leading edge of the wavefront had an abrupt slowing (the velocity around the V1/V2 border was about 5 mm/s). Meanwhile the trailing edge of the wave was still in V1 and maintained a higher speed (50 - 70 mm/s). As a result, a thin band of compressed activity formed along the V1/V2 border (Figure 5B). The compression and the resulting thin band sustained for a relatively long time compared to the time taken for the activity expanding within V1. In order to analyze the abrupt slowing of the wave, we present the data in another form of pseudo-color map, the X-T map (Figure 5C), in which the signal on a row of detectors along the propagating direction was displayed against time. In the X-T map the slope of the leading edge is proportional to the propagating velocity, and slowing of the wavefront can be identified as a reduction in the slope. Wave compression can be clearly seen as a thin horizontal stripe at the V1/V2 border (Figure 5C), indicating a nearly zero propagating velocity for about 35 ms during the course of the wave compression.

Figure 5
Velocity change during wave compression

We assumed that inhibition in local circuits may play a role in the control of velocity. To test if wave compression can be modulated by GABAA inhibition, we applied bicuculline, a GABAA receptor antagonist, to the cortex. The bicuculline was applied epi-durally with a low concentration of 3 - 5 μM, which is below the threshold of interictal-like spikes (5-10 μM). At low concentration, bicuculline can completely abolish the wave compression without significantly changing the speed of wave propagation within V1 (Figure 6), suggesting that inhibition in the local circuit plays a major role in the wave compression. Compression bands reappeared after bicuculline was washed out (data not shown), suggesting that elimination of the compression band does not require a permanent change in the cortical circuit. Under low dose of bicuculline perfusion, the propagating velocity across the V1/V2 border was same as that within V1 and V2 (Figure 6B), suggesting that changes in the excitatory connections at the border do not play a major role in the compression.

Figure 6
Bicuculline eliminates the wave compression

Origin of reflected wave

Reflected waves, while more variable, were observed in most trials (86%, 168/194) following the compression, and they originated near the compression band (Figures (Figures1,1, ,5C).5C). Since corpus callosum afferent fiber bundles are concentrated near the borders between visual areas, we wanted to determine if the reflected waves were initiated via the callosal fibers by the activity on the contralateral side of the cortex. Locally applied lidocaine or CNQX to the contralateral cortex suppressed the local EEG response on the contralateral cortex significantly but did not block the reflected wave (supplemental Figure S5), suggesting that the input from contralateral cortex is not a major contributor to the reflected wave. Thus, the reflected waves are likely to originate ipsilaterally; they may be a feedback wave from higher visual areas.

Evoked waves vs. spontaneous waves

Both evoked and spontaneous cortical activities manifested as propagating waves. The spontaneous activities have also been referred to as “UP states” (Petersen et al, 2003b). It is difficult to distinguish evoked events from spontaneous events in a recording from a single site. However, the spatiotemporal pattern of these two types of events differed markedly. Figure 7 shows wave patterns of two evoked and six spontaneous events from the same animal. The evoked waves were initiated in V1, compressed near the V1/V2 border, and had a reflected wave (Figure 7A). In contrast, the six spontaneous events all initiated from different locations and propagated across the cortex with various directions (Figure 7B). Compression and reflection were not observed during these spontaneous waves.

Figure 7
Compression only occurs in evoked waves

To further elucidate the difference between evoked and spontaneous waves, we examined a large number of spontaneous and evoked events. Figure 8A shows the distribution of initiation sites of 20 evoked events and 123 spontaneous events from one animal. The initiation sites of the evoked events were clustered in the V1 monocular area, while the spontaneous events started at various locations, many from outside of the imaged area. Since the evoked waves had compression at the V1/V2 border, their overall propagating velocity might be slower than that of spontaneous waves. The velocity of 89 evoked and 354 spontaneous events in 5 animals was examined (Figure 8B). We recorded the peak time for a wave to reach each detector and calculated the standard deviation (SD) of the peak time for all detectors (Figure 8B, left). A larger SD indicates longer delay between initiation site and other locations or a slower wave, while smaller SD indicating shorter delay or a faster wave. (This method simplifies the calculation of the velocity because propagation direction vectors can be ignored). The distribution of SDs showed that most of the evoked events had large SDs compared to the spontaneous ones (Figure 8B, right, p < 0.001, t-test); while SDs of 80% of spontaneous events were between 0 - 20ms, 80% of evoked SDs were between 20 - 40ms.

Figure 8
Comparison of evoked and spontaneous events


The principal findings of this study are: 1. Visually evoked activity in rat visual cortex manifests as a wave propagating from V1 to other visual cortical areas. 2. The evoked wave is compressed at the border between visual areas, and a reflected wave is initiated after the compression. The compression and reflection occur robustly and reproducibly in different trials and in different animals. 3. GABAergic inhibition near the border between V1 and V2 plays a major role in the wave compression. 4. The compression and reflection only occur in visually evoked waves but not during spontaneous events, suggesting that the compression/reflection pattern is governed by a mechanism associated with visual processing.

Studies on visual processing have emphasized the receptive fields of individual neurons and the input-output relationship at the single cell level. Spatiotemporal dynamics due to interactions in large networks, while important to the integration of information at the system level, are much less understood. In this report, we have observed complex and highly reproducible wave patterns, implying an internal mechanism organizing the activity at population level. Such intriguing patterns have not been reported in the previous studies of cortical waves. This is probably because wave compression/reflection pattern is not time locked to the onset of stimuli and may be blurred when averaging multiple trials. Thus, visualizing waves in single trials using blue voltage-sensitive dyes (Shoham et al., 1999) and a high dynamic range imaging apparatus (Wu and Cohen, 1993; Lippert et al., 2007) is essential for our findings.

Propagating waves in sensory cortices

In mammalian sensory cortex, sensory evoked propagating waves were found in previous imaging studies using blue dyes (Derdikman et al., 2003; Petersen et al., 2003a; Roland et al., 2006). During these waves, neurons in layer II-III depolarize for a few millivolts above the resting potential (Petersen et al., 2003b) and thus the firing probability is modulated. Multiple-peaks in VSD signal (e.g., Figure 1, traces 1- 2) during the primary/reflected waves suggest biphasic response in the spiking of individual neurons (see figure 4 of Roland et al., 2006).

As a common feature, sensory evoked waves robustly initiated from the location of cortical afferents and propagated over a large area. Due to the propagation, a time delay is spatially distributed over the cortical area as determined by the propagating velocity. On a population scale, such delayed activation is different from the synchrony on a millisecond scale between active neurons. Wave compression/reflection observed in this report suggests an even larger time delay, in that the depolarization in V2 is about ∼30 ms after V1 is activated (Figure 5C). The reflected wave, in contrast, would allow V1 and V2 to be depolarized together within 10 ms, following the compression. This distinct temporal pattern provides a mechanism for simultaneously depolarizing neurons in several visual areas. Neurons in two different visual areas may simultaneously increase their firing probability by the wave, during a particular period after receiving visual stimulus, thus facilitating the information exchange between these areas.

Feedback waves traveling from areas 21 and 19 towards area 18 and 17 were recently reported by voltage-sensitive dye imaging in ferrets (Roland et al., 2006). While marked differences in latency and propagating velocity were seen between their data and ours, in general, both forward and backward waves were observed, thus suggesting that propagating waves are common phenomena during visual processing.

Stereotypical pattern during visually evoked activity

Compression/reflection were observed in every animal, suggesting that there is a stereotypical pattern of cortical activity for processing visual information. This pattern is likely to be governed by an internal mechanism that is not activated during spontaneous events. Propagating waves are known to change velocity and/or direction due to dynamic interactions with other waves. For example, in brain slices, collision of two waves propagating toward each other results in annihilation (Wu et al., 1999) or formation of spiral waves (Huang et al., 2004). Reflection was also frequently observed in brain slices (Bao and Wu, 2003). However, wave-to-wave interactions in brain slices occurred at various locations with uncertain wave patterns (Huang et al., 2004). Such interactions are dynamic and different from the wave compression/reflection reported here, because the latter occurred at a fixed location and had similar pattern from trial to trial (Figure 2).

Wave compression is a result of sudden reduction of propagating velocity of the leading edge of the wave near the V1/V2 border (Figure 5C). We consider two possible mechanisms underlying the abrupt slowdown of the leading edge: One is a reduction in the horizontal connections near the V1/V2 border and the second is an increase in the local circuit inhibition. Abundant horizontal connections exist between the pyramid neurons in layer 2/3 of visual cortex (Gilbert and Wiesel, 1979; Rockland and Lund, 1982, 1983; Livingstone and Hubel, 1984; Martin and Whitteridge, 1984; Gilbert and Wiesel, 1989), which are thought to mediate the subthreshold activation over a large area (Das and Gilbert, 1995; Toth et al., 1996). Computational models suggest that reflected waves can also occur when a wave runs into an area with decrease excitatory interactions (Ermentrout and Rinzel, 1996). However, changes of horizontal excitatory connections near the V1/V2 border itself cannot explain the wave compression, because spontaneous waves do not slow down at the V1/V2 border (Figure 7). Dynamic increase of GABAA inhibition during the evoked activity offers another mechanism. Inhibition in cortical local circuits is known to be important for controlling propagation velocity (Traub et al., 1987; Chervin et al., 1988; Miles et al., 1988; Chagnac-Amitai and Connors, 1989; Wadman and Gutnick, 1993; Golomb and Amitai, 1997; Laaris et al., 2000; Wu et al., 2001; Golomb and Ermentrout, 2002). Indeed, we found that bicuculline completely eliminated the compression at the V1/V2 border (Figure 6), suggesting that GABAA inhibition provides a mechanism for the wave compression. Such inhibition is dynamic and temporary because it is exclusively related to visually evoked waves. Interictal-like spikes occur in visual cortex when 10-20% of GABAA inhibition is reduced (Chagnac-Amitai et al., 1989). Wave compression can be disrupted below the threshold of interictal-like spikes, suggesting that wave compression requires a delicate balance of GABAergic inhibition.

From a computational perspective, cortical neuronal populations may be viewed as loosely coupled oscillators (Grannan et al 1993). Visual stimulus may increase the interactions and change phase shift among the oscillators. When the stimulus reaches threshold, the magnitude of the interactions will be high enough to initiate the primary wave. The velocity of the propagation of the wave may be determined by the phase shift among the neuronal oscillators (Ermentrout and Kleinfeld, 2001). Our results suggest that GABAergic inhibition also increased during evoked events, causing wave compression at the border between visual areas. Apparently, spontaneous waves sustained by a different process, cortical neurons may receive non-specific and synchronized input from subcortical structures (Steriade et al., 1997), resulting in a small phase shift and a fast overall propagation velocity.

We have observed the same propagation pattern when the stimulus was drifting at various orientations. This may be due to the lack of orientation columns in rodent visual cortex, with cells responding to different stimulus orientations mixed in the V1 area (Girman et al., 1999; Ohki et al., 2005;Van Hooser et al., 2005;Yoshimura et al., 2005). The inter-columnar projections in layer II/III and light scattering in cortical tissue are potential factors that might blur the boundary of the columnar structures. However, stimulus presented at different locations in the visual field did affect the shape and location of the compression band (Supplemental Figure S3A, S3B), suggesting interactions between propagation waves and cortical columnar structures. We speculate that in species with well-developed orientation columns, the fine structure of the initiation of the wave may vary when the orientation of the visual stimulus changes.

In conclusion, we have observed a stereotypical pattern of wave compression and reflection during visually evoked cortical activity. This pattern occurs robustly during a variety of visual stimulus, but not during spontaneous events. Such patterns may provide a mechanism to simultaneously depolarize a large population of neurons across two visual areas, and may have important implications for visual processing.

Experimental Procedures

Surgical procedure

Adult Long-Evans rats (250-400g, N = 36) were used in the experiments. Surgical procedures were approved by Georgetown University Animal Care and Use Committee, strictly following NIH recommendations and guidelines.

Before surgery, the animal was given an intraperitoneal (IP) injection of atropine (60 μg/kg). Anesthesia was induced with 4 % isoflurane in air. After a tracheostomy tube was inserted, the animal was connected to a small animal respirator (Harvard Apparatus) and the concentration of isoflurane was reduced to 2.5% in pure oxygen for surgery, and 1.5%-2.0% through the imaging experiment. The respiratory rate (60-100 c/min) and volume (2-3 ml) were adjusted such that the inspiratory pressure was between 5-10 mm H2O and the end-tidal (ET) CO2 was 25 - 35 mm Hg (3.3 - 4.6 %). The body temperature of the anesthetized animals was maintained at 37°C with a regulated heating pad. A cranial window (5 × 5 mm2) was drilled over the visual cortex of the left hemisphere (bregma -4 to -9 mm, lateral 0.5- 5.5 mm). The bone was carefully separated from the dura and great care was taken to avoid irritating the dura and the cortex underneath by touching or excessive pressure. Irritated dura or cortex often led to poor staining and thus careful craniotomy was important for successful staining. In some experiments, dexamethasone sulfate (1 mg/kg IP) was given a few hours prior to the surgery to reduce inflammatory response of the dura.

Dye staining

The cortex was stained through the dura. Leaving dura intact significantly reduces the movement artifact during optical recording (London et al., 1989). In order to increase the dural permeability to the dye, we dried the dura with gentle air flow before staining. The voltage-sensitive dye RH-1691 or RH-1838 (Optical Imaging, www.opt-imaging.com) was dissolved in ringer solution (1- 2 mg/ml) and ∼200 μl dye solution was used for staining an area of 5 mm in diameter. During staining, the dye solution was continuously circulated by a perfusion pump (London et al., 1989). The pump drew a small amount (∼100 μl) of the dye solution from the top of the dura, held it for half a second and then released the drop back to the pool. Using the circulation greatly improved the staining quality. After staining for 90 min, the cortex was washed with dye-free ringer solution for ∼30 minutes. Our method provided a good staining over the cortical layer I-III (supplemental Figure S6), similar to that stained without dura (Kleinfeld and Delaney, 1996; Ferezou et al., 2006).

Optical imaging

The cortex was imaged with a 5× macroscope (Kleinfeld et al., 1994) with a field of view approximately 4 mm in diameter. Light from a tungsten filament lamp (12V, 100W, Zeiss) was filtered by a 630±15 nm interference filter and then reflected down onto the cortex via a 655 nm dichroic mirror (Chroma Technology). Köhler illumination was achieved through the macroscope. The cortex was exposed to the light only during recording trials. Dye fluorescence was filtered with a 695 nm long-pass filter and projected onto the fiber optic aperture of a 464-channel photodiode array (WuTech Instruments). Each channel (pixel) of the array receives light from a cortical area of 160 μm in diameter. The photocurrent from each channel is individually amplified with a two-stage amplifier system (Wu and Cohen, 1993). At the output of the second stage amplifier, a signal of 10-3 will span a range of 6 bits when digitized with a 12-bit A/D converter at 1.6 kHz.

Local EEG, ECG, tracheal respiratory pressure and sensor signal monitoring the visual stimulation were digitized simultaneously with the optical channels. Local EEG was recorded with a silver ball electrode placed at the corner of the imaging field, amplified 1000 times and filtered between 0.2 - 400 Hz. ECG and tracheal pressure were used for removing pulsation and respiration artifacts offline.

Subtracting brain pulsation artifact

Pulsation and respiration artifacts were time-locked to the ECG and tracheal pressure, and an algorithm was used to separate the artifacts from the signal. The algorithm was modified from our previous methods (Ma et al., 2004). Briefly, an “averaged pulsation artifact” was obtained for each optical detector. During each 5 sec recording trial, there will be ∼30 heartbeats. Since neuronal activity was not time-locked to the ECG, in the averaged pulsation artifact, the signal would be reduced ∼30 fold. Therefore ECG triggered subtraction removes the components time-locked to the ECG, but has little effect on the signal. The algorithm was implemented in Matlab (Mathworks, Natick, MA). We used NeuroPlex (RedshirtImaging, Decatur, GA) to record and view data during experiments and Matlab for data analysis and making figures.

Sensitivity of optical imaging

In order to verify the sensitivity of voltage-sensitive dye recording, we simultaneously recorded the optical signal and local field potentials from the same location in visual cortex (Supplemental Figure S1). Under isoflurane anesthesia, both spontaneous and evoked events in the local field potential were also seen in the voltage-sensitive dye signals (Supplemental Figure S1A, B). Note that almost every peak in the local EEG also occurs in the optical recordings, demonstrating that the sensitivity of our optical recording is comparable to that of local EEG recordings. This sensitivity is essential for visualizing wave compression/reflection in single trials without averaging. However, the waveform of the EEG and optical recordings are not exactly the same, probably because the local EEG electrode picked up signals from strong current source in deep cortical layers or subcortical structures while voltage-sensitive dye signal was localized to the neurons in cortical layer I-III under each optical detector.


Visual stimulation patterns were generated by programs written in Visual C++. The patterns were displayed by a screen projector, projecting to a screen of 10 × 7 inches. The resolution of the projector was 1024 × 768 with a refresh rate of 60 Hz. The screen was placed approximately 20 cm in front of the animal’s contralateral eye (Supplemental Figure S2). The visual stimulus presented to the contralateral eye cannot be seen by the ipsilateral eye and so that the ipsilateral eye was not covered in the most of the experiments. A sinusoidal grating (0.025 - 0.3 cycles/degree, 50w×38h degrees of viewing angle) was constantly presented to the contralateral eye. The stimulation is the drifting of the grating. The stimulation duration was 500∼2000 ms and the velocity of the drifting was 30 - 200 degree/sec. The visual stimulation was monitored by a photosensor attached to the corner of the screen. The output of the sensor was digitized simultaneously with the imaging data.

Data analysis and pseudo-color images

Data analysis was done with scripts written with Matlab (Mathworks). The pseudo-color images and movies were generated from the fractional changes of the fluorescent light on a linear color scale. Briefly, during data acquisition, the resting fluorescent light on each detector was removed by the amplifier hardware and the fractional changes in the fluorescence was amplified and digitized. In data analysis the fractional change in light on each detector was normalized between pre-stimulus baseline and the peak of the primary wave. The normalized value was assigned to colors (red = 1 to blue = 0) according to a linear pseudo-color scale (Grinvald et al., 1982; Jin et al., 2002; Ma et al., 2004).

Supplementary Material



Supplemental movie 1 Compression band and reflected wave

The field of view is the same as that in Figure 5B, about 4 mm in diameter with the V1/V2 border approximately at the center of the field. The play speed of the movie is 100-fold slower than the real event. The color scale is linear and the full range of the scale (red--blue) is about 0.1% of the fractional change in the fluorescent intensity. The primary wave is initiated at the upper-left corner of the imaging field. The first wave compression occurs near the center of the imaging field, followed by a reflecting wave. Note that the second compression and the consequent reflecting wave occurred at the right edge of the imaging field.


Supplemental movie 2. Wave compression/reflection is robust between trials

The three movies are from three trials of the same imaging field with identical stimuli. The movies are played simultaneously to show the comparison of the spatiotemporal patterns of the compression/reflection. The playing speed is 100-fold slower than the real events. The imaging field is about 4 mm in diameter.


We thank Drs. L.B. Cohen, G.B. Ermentrout, S.J. Schiff, S. Vicini, B. Tian and E. Galloway for helpful discussions. Supported by NIH grant NS36447(JYW), the American Epilepsy Society and the Lennox Trust Fund (XH).


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Angelucci A, Levitt JB, Walton EJ, Hupe JM, Bullier J, Lund JS. Circuits for local and global signal integration in primary visual cortex. J Neurosci. 2002;22:8633–8646. [PubMed]
  • Arieli A, Shoham D, Hildesheim R, Grinvald A. Coherent spatiotemporal patterns of ongoing activity revealed by real-time optical imaging coupled with single-unit recording in the cat visual cortex. J Neurophysiol. 1995;73:2072–2093. [PubMed]
  • Bao W, Wu JY. Propagating wave and irregular dynamics: spatiotemporal patterns of cholinergic theta oscillations in neocortex in vitro. J Neurophysiol. 2003;90:333–341. [PMC free article] [PubMed]
  • Chagnac-Amitai Y, Connors BW. Horizontal spread of synchronized activity in neocortex and its control by GABA-mediated inhibition. J Neurophysiol. 1989;61:747–758. [PubMed]
  • Chen Y, Geisler WS, Seidemann E. Optimal decoding of correlated neural population responses in the primate visual cortex. Nat Neurosci. 2006;9:1412–1420. [PMC free article] [PubMed]
  • Chervin RD, Pierce PA, Connors BW. Periodicity and directionality in the propagation of epileptiform discharges across neocortex. J Neurophysiol. 1988;60:1695–1713. [PubMed]
  • Civillico EF, Contreras D. Integration of evoked responses in supragranular cortex studied with optical recordings in vivo. J Neurophysiol. 2006;96:336–351. [PubMed]
  • Das A, Gilbert CD. Long-range horizontal connections and their role in cortical reorganization revealed by optical recording of cat primary visual cortex. Nature. 1995;375:780–784. [PubMed]
  • Delaney KR, Gelperin A, Fee MS, Flores JA, Gervais R, Tank DW, Kleinfeld D. Waves and stimulus-modulated dynamics in an oscillating olfactory network. Proc Natl Acad Sci U S A. 1994;91:669–673. [PMC free article] [PubMed]
  • Derdikman D, Hildesheim R, Ahissar E, Arieli A, Grinvald A. Imaging spatiotemporal dynamics of surround inhibition in the barrels somatosensory cortex. J Neurosci. 2003;23:3100–3105. [PubMed]
  • Ermentrout GB, Kleinfeld D. Traveling electrical waves in cortex: insights from phase dynamics and speculation on a computational role. Neuron. 2001;29:33–44. [PubMed]
  • Ermentrout GB, Rinzel J. Reflected waves in an inhomogeneous excitable medium. SIAM J Appl Math. 1996;56:1107–1128.
  • Ferezou I, Bolea S, Petersen CC. Visualizing the cortical representation of whisker touch: voltage-sensitive dye imaging in freely moving mice. Neuron. 2006;50:617–629. [PubMed]
  • Freeman WJ, Barrie JM. Analysis of spatial patterns of phase in neocortical gamma EEGs in rabbit. J Neurophysiol. 2000;84:1266–1278. [PubMed]
  • Girman SV, Sauve Y, Lund RD. Receptive field properties of single neurons in rat primary visual cortex. J Neurophysiol. 1999;82:301–311. [PubMed]
  • Gilbert CD, Wiesel TN. Morphology and intracortical projections of functionally characterised neurones in the cat visual cortex. Nature. 1979;280:120–125. [PubMed]
  • Gilbert CD, Wiesel TN. Columnar specificity of intrinsic horizontal and corticocortical connections in cat visual cortex. J Neurosci. 1989;9:2432–2442. [PubMed]
  • Golomb D, Amitai Y. Propagating neuronal discharges in neocortical slices: computational and experimental study. J Neurophysiol. 1997;78:1199–1211. [PubMed]
  • Golomb D, Ermentrout GB. Slow excitation supports propagation of slow pulses in networks of excitatory and inhibitory populations. Phys Rev E Stat Nonlin Soft Matter Phys. 2002;65:061911. [PubMed]
  • Grannan ER, Kleinfeld D. Stimulus-dependent synchronization of neuronal assemblies. Neural Comput. 1993;5:550–569.
  • Grinvald A, Hildesheim R. VSDI: a new era in functional imaging of cortical dynamics. Nat Rev Neurosci. 2004;5:874–885. [PubMed]
  • Grinvald A, Manker A, Segal M. Visualization of the spread of electrical activity in rat hippocampal slices by voltage-sensitive optical probes. J Physiol. 1982;333:269–291. [PMC free article] [PubMed]
  • Huang X, Troy WC, Yang Q, Ma H, Laing CR, Schiff SJ, Wu JY. Spiral waves in disinhibited mammalian neocortex. J Neurosci. 2004;24:9897–9902. [PubMed]
  • Jin W, Zhang RJ, Wu JY. Voltage-sensitive dye imaging of population neuronal activity in cortical tissue. J Neurosci Methods. 2002;115:13–27. [PubMed]
  • Kennedy H, Bullier J. A double-labeling investigation of the afferent connectivity to cortical areas V1 and V2 of the macaque monkey. J Neurosci. 1985;5:2815–2830. [PubMed]
  • Kleinfeld D, Delaney KR, Fee MS, Flores JA, Tank DW, Gelperin A. Dynamics of propagating waves in the olfactory network of a terrestrial mollusk: an electrical and optical study. J Neurophysiol. 1994;72:1402–1419. [PubMed]
  • Kleinfeld D, Delaney KR. Distributed representation of vibrissa movement in the upper layers of somatosensory cortex revealed with voltage-sensitive dyes. J Comp Neurol. 1996;375:89–108. [PubMed]
  • Laaris N, Carlson GC, Keller A. Thalamic-evoked synaptic interactions in barrel cortex revealed by optical imaging. J Neurosci. 2000;20:1529–1537. [PubMed]
  • Lam YW, Cohen LB, Wachowiak M, Zochowski MR. Odors elicit three different oscillations in the turtle olfactory bulb. J Neurosci. 2000;20:749–762. [PubMed]
  • Lam YW, Cohen LB, Zochowski MR. Odorant specificity of three oscillations and the DC signal in the turtle olfactory bulb. Eur J Neurosci. 2003;17:436–446. [PubMed]
  • Lippert MT, Takagaki K, Xu W, Huang X, Wu JY. Methods for voltage-sensitive dye imaging of rat cortical activity with high signal-to-noise ratio. J Neurophysiol. 2007 [PMC free article] [PubMed]
  • Livingstone MS, Hubel DH. Specificity of intrinsic connections in primate primary visual cortex. J Neurosci. 1984;4:2830–2835. [PubMed]
  • Livingstone MS, Hubel DH. Connections between layer 4B of area 17 and the thick cytochrome oxidase stripes of area 18 in the squirrel monkey. J Neurosci. 1987;7:3371–3377. [PubMed]
  • Livingstone MS, Hubel DH. Segregation of form, color, movement, and depth: anatomy, physiology, and perception. Science. 1988;240:740–749. [PubMed]
  • London JA, Cohen LB, Wu JY. Optical recordings of the cortical response to whisker stimulation before and after the addition of an epileptogenic agent. J Neurosci. 1989;9:2182–2190. [PubMed]
  • Ma HT, Wu CH, Wu JY. Initiation of spontaneous epileptiform events in the rat neocortex in vivo. J Neurophysiol. 2004;91:934–945. [PMC free article] [PubMed]
  • Martin KA, Whitteridge D. Form, function and intracortical projections of spiny neurones in the striate visual cortex of the cat. J Physiol. 1984;353:463–504. [PMC free article] [PubMed]
  • Miles R, Traub RD, Wong RK. Spread of synchronous firing in longitudinal slices from the CA3 region of the hippocampus. J Neurophysiol. 1988;60:1481–1496. [PubMed]
  • Ohki K, Chung S, Ch’ng YH, Kara P, Reid RC. Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature. 2005;433:597–603. [PubMed]
  • Olavarria JF, Hiroi R. Retinal influences specify cortico-cortical maps by postnatal day six in rats and mice. J Comp Neurol. 2003;459:156–172. [PubMed]
  • Paxinos G, Watson C. The rat brain in stereotaxic coordinates. fifth edition (Elsevier academic press); 2005.
  • Petersen CC, Grinvald A, Sakmann B. Spatiotemporal dynamics of sensory responses in layer 2/3 of rat barrel cortex measured in vivo by voltage-sensitive dye imaging combined with whole-cell voltage recordings and neuron reconstructions. J Neurosci. 2003a;23:1298–1309. [PubMed]
  • Petersen CC, Hahn TT, Mehta M, Grinvald A, Sakmann B. Interaction of sensory responses with spontaneous depolarization in layer 2/3 barrel cortex. Proc Natl Acad Sci U S A. 2003b;100:13638–13643. [PMC free article] [PubMed]
  • Prechtl JC, Cohen LB, Pesaran B, Mitra PP, Kleinfeld D. Visual stimuli induce waves of electrical activity in turtle cortex. Proc Natl Acad Sci U S A. 1997;94:7621–7626. [PMC free article] [PubMed]
  • Prechtl JC, Bullock TH, Kleinfeld D. Direct evidence for local oscillatory current sources and intracortical phase gradients in turtle visual cortex. Proc Natl Acad Sci U S A. 2000;97:877–882. [PMC free article] [PubMed]
  • Rockland KS, Pandya DN. Cortical connections of the occipital lobe in the rhesus monkey: interconnections between areas 17, 18, 19 and the superior temporal sulcus. Brain Res. 1981;212:249–270. [PubMed]
  • Rockland KS, Lund JS. Widespread periodic intrinsic connections in the tree shrew visual cortex. Science. 1982;215:1532–1534. [PubMed]
  • Rockland KS, Lund JS. Intrinsic laminar lattice connections in primate visual cortex. J Comp Neurol. 1983;216:303–318. [PubMed]
  • Roland PE, Hanazawa A, Undeman C, Eriksson D, Tompa T, Nakamura H, Valentiniene S, Ahmed B. Cortical feedback depolarization waves: A mechanism of top-down influence on early visual areas. Proc Natl Acad Sci U S A. 2006;103:12586–12591. [PMC free article] [PubMed]
  • Rubino D, Robbins KA, Hatsopoulos NG. Propagating waves mediate information transfer in the motor cortex. Nat Neurosci. 2006;9:1549–1557. [PubMed]
  • Senseman DM, Robbins KA. Modal behavior of cortical neural networks during visual processing. J Neurosci. 1999;19:RC3. [PubMed]
  • Shmuel A, Korman M, Sterkin A, Harel M, Ullman S, Malach R, Grinvald A. Retinotopic axis specificity and selective clustering of feedback projections from V2 to V1 in the owl monkey. J Neurosci. 2005;25:2117–2131. [PubMed]
  • Shoham D, Glaser DE, Arieli A, Kenet T, Wijnbergen C, Toledo Y, Hildesheim R, Grinvald A. Imaging cortical dynamics at high spatial and temporal resolution with novel blue voltage-sensitive dyes. Neuron. 1999;24:791–802. [PubMed]
  • Sincich LC, Horton JC. Pale cytochrome oxidase stripes in V2 receive the richest projection from macaque striate cortex. J Comp Neurol. 2002a;447:18–33. [PubMed]
  • Sincich LC, Horton JC. Divided by cytochrome oxidase: a map of the projections from V1 to V2 in macaques. Science. 2002b;295:1734–1737. [PubMed]
  • Sincich LC, Horton JC. Independent projection streams from macaque striate cortex to the second visual area and middle temporal area. J Neurosci. 2003;23:5684–5692. [PubMed]
  • Steriade M. Synchronized activities of coupled oscillators in the cerebral cortex and thalamus at different levels of vigilance. Cereb Cortex. 1997;7:583–604. [PubMed]
  • Toth LJ, Rao SC, Kim DS, Somers D, Sur M. Subthreshold facilitation and suppression in primary visual cortex revealed by intrinsic signal imaging. Proc Natl Acad Sci U S A. 1996;93:9869–9874. [PMC free article] [PubMed]
  • Traub RD, Knowles WD, Miles R, Wong RK. Models of the cellular mechanism underlying propagation of epileptiform activity in the CA2-CA3 region of the hippocampal slice. Neuroscience. 1987;21:457–470. [PubMed]
  • Van Hooser SD, Heimel JA, Chung S, Nelson SB, Toth LJ. Orientation selectivity without orientation maps in visual cortex of a highly visual mammal. J Neurosci. 2005;25:19–28. [PubMed]
  • Wadman WJ, Gutnick MJ. Non-uniform propagation of epileptiform discharge in brain slices of rat neocortex. Neuroscience. 1993;52:255–262. [PubMed]
  • Wu JY, Cohen LB. Fast multisite optical measurement of membrane potential. In: Mason WT, editor. Biological Techniques: Fluorescent and Luminescent Probes for Biological Activity. Academic press; New York: 1993. pp. 389–404.
  • Wu JY, Guan L, Tsau Y. Propagating activation during oscillations and evoked responses in neocortical slices. J Neurosci. 1999;19:5005–5015. [PubMed]
  • Wu JY, Guan L, Bai L, Yang Q. Spatiotemporal properties of an evoked population activity in rat sensory cortical slices. J Neurophysiol. 2001;86:2461–2474. [PubMed]
  • Yoshimura Y, Dantzker JL, Callaway EM. Excitatory cortical neurons form fine-scale functional networks. Nature. 2005;433:868–873. [PubMed]
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Cited in Books
    Cited in Books
    PubMed Central articles cited in books
  • Compound
    PubChem Compound links
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...