Abstract

Following circumscribed retinal damage, extensive reorganization of topographically organized visual cortical areas has been demonstrated in several species of mammals (including humans). Although reorganization is often studied over extended time scales, neural response properties change within seconds of retinal deafferentation. Understanding the mechanisms underlying these short-term effects is essential for developing a complete picture of representational plasticity. One approach to the study of short-term plasticity has been to use an artificial scotoma, a stimulus-induced analog of a retinal scotoma, as a model. Here, we use event-related potentials in an artificial scotoma paradigm to examine 2 aspects of short-term plasticity in the human visual system. First, we investigated the changes within visual representations temporarily deprived of patterned visual input by probing the inner boundaries of an artificial scotoma. We found an enhanced early sensory P1, consistent with a reduction in inhibition (disinhibition), a proposed mechanism of short-term visual plasticity. Second, we investigated mechanisms through which representations of surrounding space invade a visually deprived area by probing the outer boundaries of an artificial scotoma. In this case, a later visual component, the N1, was enhanced, suggesting that feedback may provide a source of unmasked, or invading, activity to visually deprived representations.

Introduction

The topography of mammalian cortical visual areas is capable of considerable representational plasticity even in adulthood. Single-cell studies in monkeys and cats have reported extensive reorganization of visual topographic maps following induced retinal lesions. In general, these studies have found that, following several months of recovery, visually deafferented cortical areas remap their spatial representations and become responsive to regions of space well beyond their classic retinotopic boundaries (for reviews, see Gilbert 1998; Calford 2002). Though there are certain caveats (Waleszczyk et al. 2003) and reservations (e.g., Smirnakis et al. 2005) concerning topographic reorganization, there is substantial evidence of robust plasticity of spatial representations within adult mammalian visual cortex (Kaas et al. 1990; Heinen and Skavenski 1991; Gilbert and Wiesel 1992; Chino et al. 1995; Calford et al. 2000).

Evidence of topographical reorganization has been demonstrated in the human visual system as well. Psychophysical studies of patients with compromised visual input have demonstrated that scotomas are perceptually filled-in by surrounding spatial information (Zur and Ullman 2003). Furthermore, intact regions of the visual field surrounding a scotoma exhibit distortions of visual space consistent with the invasion of surrounding visual representations into deafferented visual areas (Dilks et al. 2007). Functional magnetic resonance imaging (fMRI) studies in patients with macular degeneration have further demonstrated that deafferented cortical areas become responsive to new regions of space, suggesting the occurrence of large-scale visual reorganization (Baker et al. 2005, 2008; Dilks et al. 2007, 2009; Schumacher et al. 2008; however, see Sunness et al. 2004).

Studies of representational plasticity in animals and humans have concentrated largely on the enduring topographic changes that accompany visual reorganization over extended time scales of weeks, months, or years (long-term plasticity). However, robust functional changes have been demonstrated within seconds to minutes of retinal deafferentation (short-term plasticity). Single-cell studies indicate that visual representations begin to remap almost immediately after the induction of a retinal scotoma (Heinen and Skavenski 1991; Chino et al. 1992; Gilbert and Wiesel 1992; Darian-Smith and Gilbert 1995; Schmid et al. 1995; Calford et al. 1999). Specifically, receptive fields within deafferented cortex exhibit a spatial profile up to 10 times larger than that of their classical size. Changes in receptive field dynamics over such short time scales cannot be accounted for by any significant neurite growth but must result from dynamic functional modulation of receptive field properties. Understanding the mechanisms and dynamics of short-term plasticity is critical to understanding the process of cortical reorganization and has further implications for understanding the neural substrates of perceptual filling-in (Tremere et al. 2003, 2005). Furthermore, to date, mechanisms of short-term visual plasticity remain unexplored in the human visual system.

Retinal lesioning procedures are impractical in the study of short-term plasticity and preclude its study in the human brain. A number of research groups have used a stimulus-induced analog of a retinal scotoma, an artificial scotoma, as a model to study the short-term effects of retinal deafferentation (for review, see Dreher et al. 2001). An artificial scotoma can be induced by superimposing a uniform area upon a dynamically changing textured background (Fig. 1; Ramachandran and Gregory 1991). After a short period of viewing (seconds), such a display causes the masked region of space to become filled-in by the background (Ramachandran and Gregory 1991; De Weerd et al. 1998;,De Weerd 2006). An artificial scotoma is useful in the study of short-term plasticity in that it temporarily mimics the effects of a true retinal scotoma by restricting patterned visual input from a circumscribed area of the visual field. Unlike lesioning approaches, artificial scotomas permit changes within intra-scotoma space to be examined over short time scales. Furthermore, the effects of an artificial scotoma are temporary and reversible, allowing for multiple measurements, and making it safe and practical for use in human studies. Single-cell studies in cats and monkeys have demonstrated that the short-term effects of an artificial scotoma are similar to that of a true retinal lesion. Receptive fields within artificial scotoma boundaries exhibit enlarged spatial profiles and increased response gain following a period of conditioning with an artificial scotoma (Pettet and Gilbert 1992; Das and Gilbert 1995b; DeAngelis et al. 1995; Volchan and Gilbert 1995). Effects of plasticity induced by artificial scotomas have been documented in striate cortex as well as extrastriate visual areas (Pettet and Gilbert 1992; Das and Gilbert 1995b; De Weerd et al. 1995).

Figure 1.

A static depiction of an artificial scotoma display. The perceptual effect can be observed, even in a static display, if one fixates the star in the lower left corner. After several seconds of viewing, the gray circle will appear to fade from view, becoming filled-in by the textured background.

Figure 1.

A static depiction of an artificial scotoma display. The perceptual effect can be observed, even in a static display, if one fixates the star in the lower left corner. After several seconds of viewing, the gray circle will appear to fade from view, becoming filled-in by the textured background.

Considerable evidence indicates that, for both short- and long-term plasticity, visually deprived cortical representations become invaded by surrounding representations through a process of disinhibition (Pettet and Gilbert 1992; Darian-Smith and Gilbert 1994, 1995; Das and Gilbert 1995a, 1995b; DeAngelis et al. 1995; De Weerd et al. 1995; Rosier et al. 1995; Eysel et al. 1999; Arckens et al. 2000). Disinhibition refers to a reduction in local inhibition that, in turn, “unmasks” subthreshold excitatory inputs from surrounding spatial representations, allowing the activity of these inputs to invade visually deprived cortical representations. In the case of long-term visual reorganization, invading input appears to be mediated largely by local long-range horizontal connections (Darian-Smith and Gilbert 1994, 1995; Calford et al. 2003). Though feedback from later visual areas does not appear to account substantially for long-term topographical reorganization in cats or monkeys (Chino 1995; Darian-Smith and Gilbert 1995; Young et al. 2002), neuroimaging results indicate that it may play a more significant role in visual reorganization in the human brain (Masuda et al. 2008). Furthermore, the rapid dynamic changes in visual responses that occur over the short term may be driven, at least in part, by feedback from later stages of visual processing. Receptive field characteristics can be dramatically modulated by feedback connections (Angelucci and Bullier 2003), but the involvement of feedback projections from later visual areas remains an unexplored possibility in the short-term plasticity literature.

Here, we used event-related potentials (ERPs) to examine short-term visual plasticity in humans using the artificial scotoma as a model of short-term plasticity. Artificial scotomas were conditioned for a 6-s period after which time a visual probe was flashed. In 2 experiments, we examined 2 aspects of short-term plasticity by probing different regions of scotoma space. In Experiment 1, we measured ERPs in response to probes presented within the inner boundaries of an artificial scotoma (scotoma zone; Fig. 2). Scotoma-zone ERPs reflect the cortical response of those visual representations deprived of visual input by an artificial scotoma. We hypothesized that disinhibition would lead to an increased visual cortical response that would be measurable in the early sensory components as increased amplitude (relative to a control condition). Specifically, we evaluated the P1 (120–150 ms) and N1 (150–220 ms) components for such modulations, as these components reflect activity within extrastriate visual cortical areas (Di Russo et al. 2001). In Experiment 2, we measured ERPs from the region of space surrounding the outer boundaries of an artificial scotoma (border zone; Fig. 2). Border-zone ERPs reflect the cortical response of representations of visual space adjacent to the artificial scotoma. Disinhibition is thought to unmask excitatory inputs from cortical representations of space adjacent to those of a scotoma. We hypothesized that border-zone ERPs would provide an index of this invading activity. That is, the additional excitatory input to scotoma representations revealed by disinhibition should result in a measurable change in the amplitude of sensory ERP components. Again, P1 and N1 components were the focus of our investigation. In addition to providing an index of unmasked activity in humans, we also aimed to investigate the involvement of feedback from later visual areas. We reasoned that if unmasked excitatory activity originates locally, then it should be reflected in the same ERP component(s) as those of the scotoma-zone probes. However, if this information arises from a later cortical source, we should expect to see it manifest in later components of the ERP.

Figure 2.

Illustration of scotoma-zone and border-zone regions.

Figure 2.

Illustration of scotoma-zone and border-zone regions.

Materials and Methods

Subjects

Subjects were recruited from the Georgia Institute of Technology undergraduate population, reported normal or corrected to normal vision, and gave informed consent prior to participation. Twenty-one subjects participated in Experiment 1 (10 females), ranging in age from 18 to 22 (Mean [M] = 19.5, standard deviation [SD] = 1.0). Three of these subjects were subsequently dropped due to poor signal-to-noise ratio of ERPs. Fifteen subjects were recruited for Experiment 2 (8 females), ranging in age from 18 to 21 (M = 20, SD = 1.2). Subjects in each experiment participated in 2 sessions conducted on separate days.

Stimuli and Procedure

Experimentation was conducted in a sound-attenuating chamber under low levels of ambient illumination. Stimuli were presented on a 21-inch CRT monitor using the Presentation stimulus package (Neurobehavioral Systems, Albany, CA). Monitor gamma was corrected in software to a value of 1.0 (linear) and verified with a photometer. A viewing distance of 57 cm was maintained using a chin rest.

The general sequence of events for experimental trials consisted of a 6 s exposure to a dynamic background stimulus (conditioning phase) followed by the presentation of a visual probe. The introduction and duration of artificial scotomas during the conditioning phase were manipulated in 2 experimental conditions (Scotoma and Sham conditions). Experiments 1 and 2 varied only the region of space being stimulated by the visual probe; all other experimental parameters were identical between the 2 experiments. The inner boundaries of an artificial scotoma (scotoma zone) were probed in Experiment 1, whereas the outer boundaries (border zone) were probed in Experiment 2.

The conditioning phase consisted of a dynamic background of 1200 small (0.1° × 1.0°), high luminance (137.0 cd/m2), vertically oriented bars. These bars were present in random positions on a dark background (33.6 cd/m2) with a red fixation dot (0.2° × 0.2°) in the center of the display. Each vertical bar was randomly repositioned within a 36° × 26° area every 50 ms (20 Hz) with the restriction that a bar could not be positioned such that it overlapped in space with the fixation dot. Artificial scotoma stimuli were uniform gray discs (75.1 cd/m2; 2.0° diameter) superimposed upon the dynamic background. Scotoma discs were centered 5.0° from fixation and were always presented as a pair such that a single disc was present in the left and right visual field. The size and eccentricity of scotoma discs were chosen based on previous studies in humans and monkeys indicating that, on average, these parameters will induce the predicted perceptual and neurophysiological effects within a 6-s conditioning period (De Weerd et al. 1995, 1998).

There were 2 manipulations of theoretical interest for each experiment. In one condition (the Scotoma condition), an artificial scotoma was induced by superimposing the scotoma discs on the dynamic background for the duration of the conditioning phase (i.e., 6 s). In a second condition (the Sham condition), the scotoma discs were introduced only for the last 1 s of the conditioning phase. That is, the dynamic background stimuli modulated for a period of 5 s after which time the scotoma discs were superimposed upon the dynamic background for the remaining 1 s of the conditioning phase. The plasticity introduced by an artificial scotoma requires several seconds of exposure to develop (De Weerd et al. 1995). Thus, neurophysiological consequences of an artificial scotoma should be largest and most pronounced in the Scotoma condition where the scotoma is conditioned for an extensive period of time. The Sham condition serves as a stimulus-matched control since the stimulus parameters are identical with the exception of the duration of scotoma conditioning.

Five hundred milli seconds prior to the end of the conditioning period, a warning stimulus appeared. The warning stimulus consisted of a brief (50 ms) color and luminance change in the fixation dot (from red to white). The conditioning phase terminated with the removal of the dynamic background, leaving only the scotoma discs and a fixation point. This display persisted for a random interval between 300 and 700 ms after which time the visual probe flashed for 50 ms within the boundaries of 1 of the 2 scotoma discs. Visual probes were grating annuli that consisted of a 5 cpd sinusoidal annulus centered on a Gaussian envelope. Annuli were tilted either clockwise or anticlockwise tilt by 3°. Sinusoidal phase of grating annuli varied randomly between 0°, 30°, 60°, 90°, 120°, and 150° on a trial-to-trial basis. In Experiment 1, annuli were positioned within the scotoma zone, abutting the inner boundaries of the scotoma disc (Fig. 3). Annuli in Experiment 2 surrounded the outer boundaries of the scotoma disc thus probing the border zone (Fig. 3). Probe stimuli were configured as annuli selected for 2 reasons. First, effects of short-term plasticity within the scotoma zone have been previously reported to be most pronounced nearest the inner boundaries of the scotoma (Pettet and Gilbert 1992; Das and Gilbert 1995b; De Weerd et al. 1995). Second, the use of annuli probes for Experiments 1 and 2 facilitates comparison and interpretation of results. Following the presentation of the probe stimulus, a random interval of 300–700 ms passed after which a question mark (“?”) in white text prompted observers for a response. Subjects were tasked with discriminating the tilt of the grating annulus (clockwise or anticlockwise).

Figure 3.

Sequence of event for Scotoma and Sham conditions. In the Scotoma condition, the 2 gray circles (scotoma discs) were superimposed upon the dynamically refreshing background of lines for the entire 6-s conditioning period. In the Sham condition, the scotoma discs appeared only for the last 1 s of the conditioning period. Note that all stimulus parameters were identical for the last second of the display and at the time of visual probe presentation.

Figure 3.

Sequence of event for Scotoma and Sham conditions. In the Scotoma condition, the 2 gray circles (scotoma discs) were superimposed upon the dynamically refreshing background of lines for the entire 6-s conditioning period. In the Sham condition, the scotoma discs appeared only for the last 1 s of the conditioning period. Note that all stimulus parameters were identical for the last second of the display and at the time of visual probe presentation.

An additional trial type was also included in the design to obtain a baseline visual evoked potential (VEP). This measure consisted of a brief exposure to the dynamic background stimuli (1000 ms) followed by a 500-ms period of a display consisting of the fixation and scotoma discs on the uniform background. A warning stimulus was then presented for 50 ms, and a random period between 800 and 1200 ms passed. A grating annulus probe then flashed for 50 ms. Baseline VEPs were obtained for both experiments and were used to select electrodes and time windows for statistical analysis of Scotoma and Sham conditions.

A total of 1344 trials were collected across 2 experimental sessions. Trials were blocked by condition and further subblocked by the visual field of scotoma conditioning (upper vs. lower). The sequence of blocks was counterbalanced between subjects. Between blocks of trials, the position of scotoma discs alternated between the upper and the lower halves of the visual field in order to eliminate any potential carryover effects from a previous block. That is, alternating the position of scotomas ensured that different spatial locations were conditioned and probed on each block. Thus, any residual effects of an artificial scotoma would not influence visual responses obtained in the Sham condition. A 1-min conditioning period preceded the first block of each condition.

Electrophysiological Recording and Analysis

Scalp-recorded electroencephalography (EEG) was sampled at 1024 Hz from 34 Ag–AgCl electrodes using the Active Two amplifier system (BioSemi, Amsterdam, the Netherlands). Electrodes were placed in the following positions according to a modified 10-20 system. Standard 10-20 positions were: FP1, FP2, F7, F8, F3, F4, Fz, C3, C4, Cz, P7, P8, P3, P4, Pz, T7, T8, O1, Oz, O2, M1, and M2. Additional 10-10 positions were: AF3, AF4, FC1, FC2, FC5, FC6, CP1, CP2, CP5, CP6, PO3, and PO4. Four additional electrodes were placed above and below the left eye and on the outer canthi of the left and right eyes. These leads were used to form bipolar channels for vertical electrooculogram (VEOG) and horizontal electrooculogram (HEOG), respectively.

Off-line data were rereferenced to the average electrode and digitally band-pass filtered from 0.1 to 30 Hz (24 dB/oct). Stimulus onsets were corrected for timing according to the vertical raster of the monitor. EEG data were epoched into segments of 600 ms, beginning 100 ms before stimulus onset and persisting for 400 ms thereafter. Individual segments were baseline corrected by setting the average of the 100 ms prestimulus interval to zero. Segments were considered artifacts and rejected from analysis if activity in any scalp or EOG channel exceeded ±100 μV. Segmented data were averaged by condition. Averages for probes presented in the left and right visual fields were collapsed together by respective averaging of electrodes contralateral and ipsilateral to the hemifield of probe presentation.

A posterior contralateral electrode (P7/8) was selected for statistical analysis. Two visual sensory components of the ERP (P1 and N1) were chosen a priori for statistical analysis. For both experiments, P1 amplitude was quantified as the average across a 120–150 ms time window and N1 as the average over a 190–220 ms window. Time windows were selected based on ERPs obtained from a baseline VEP (described above). A repeated-measures analysis of variance with a single factor of condition (Scotoma or Sham) was used to compare ERP amplitudes for P1 and N1 components in both Experiments 1 and 2.

Results

ERP waveforms and voltage maps for Experiments 1 and 2 are plotted in Figure 4. Mean amplitude and variability of P1 and N1 components from each experiment are given in Table 1 and Figure 5.

Table 1

Means and standard errors of P1 and N1 components (μV)

 Experiment 1 Experiment 2 
 P1 N1 P1 N1 
Condition M SE M SE M SE M SE 
Scotoma 0.56 0.11 −2.77 0.46 1.22 0.24 −3.37 0.67 
Sham 0.28 0.12 −2.58 0.44 1.500 0.32 −2.57 0.66 
 Experiment 1 Experiment 2 
 P1 N1 P1 N1 
Condition M SE M SE M SE M SE 
Scotoma 0.56 0.11 −2.77 0.46 1.22 0.24 −3.37 0.67 
Sham 0.28 0.12 −2.58 0.44 1.500 0.32 −2.57 0.66 
Figure 4.

ERP waveforms comparing Scotoma and Sham conditions from Experiments 1 and 2. Waveforms are collapsed across left and right visual fields and displayed at electrode P7/8 contralateral to visual field of probe presentation. Scalp distributions are an average of Scotoma and Sham conditions and are arbitrarily coded such that the right represents the contralateral hemisphere and left represents ipsilateral. Scotoma-zone probes from Experiment 1 revealed a significant P1 difference. Border-zone probes of Experiment 2 revealed a significant N1 effect. Single asterisk indicates significance at a level of 0.05, whereas double asterisks denote significance at a level of 0.01 or lower.

Figure 4.

ERP waveforms comparing Scotoma and Sham conditions from Experiments 1 and 2. Waveforms are collapsed across left and right visual fields and displayed at electrode P7/8 contralateral to visual field of probe presentation. Scalp distributions are an average of Scotoma and Sham conditions and are arbitrarily coded such that the right represents the contralateral hemisphere and left represents ipsilateral. Scotoma-zone probes from Experiment 1 revealed a significant P1 difference. Border-zone probes of Experiment 2 revealed a significant N1 effect. Single asterisk indicates significance at a level of 0.05, whereas double asterisks denote significance at a level of 0.01 or lower.

Figure 5.

P1 and N1 amplitude differences (Scotoma–Sham) for each subject in Experiments 1 and 2.

Figure 5.

P1 and N1 amplitude differences (Scotoma–Sham) for each subject in Experiments 1 and 2.

Perceptual performance (tilt discrimination) of Experiment 1 scotoma-zone probe stimuli was significantly worse in the Scotoma condition (M = 0.91, standard error [SE] = 0.01) relative to that of the Sham condition (F1,17 = 4.67, P < 0.05). ERP analyses revealed a significant effect in the P1 component (F1,17 = 7.75, P < 0.02) such that amplitude was greater in the Scotoma compared with Sham (Table 1). Fifteen of 18 subjects exhibited differences in this direction (Fig. 5). No significant effects were present in the N1 component (F1,17 = 1.07, P > 0.31).

Experiment 2 revealed no significant differences in tilt discrimination for border-zone probes (F1,14 = 1.93, P > 0.18; Scotoma: M = 0.87, SE = 0.02; Sham: M = 0.88, SE = 0.02). Unlike scotoma-zone probes, border-zone probes from Experiment 2 did not exhibit significant differences in P1 amplitude (F1,14 = 1.92, P > 0.18). However, N1 amplitude was significantly modulated (F1,14 = 19.45, P < 0.001), being more negative in the Scotoma condition relative to Sham (Table 1; Fig. 4). Fourteen of 15 subjects exhibited effects in this direction (Fig. 5).

Discussion

We used ERP to investigate the neural mechanisms of plasticity in the human visual system using the artificial scotoma as a model of short-term visual plasticity. Artificial scotomas were conditioned for a period of 6 s after which time ERPs were measured in response to a visual probe. We investigated 2 aspects of short-term plasticity by eliciting visual cortical responses from within the boundaries of an artificial scotoma (scotoma zone) or from the region of space surrounding an artificial scotoma (border zone). Scotoma-zone probes elicited responses from cortical representations of intra-scotoma space and index changes in visual response properties due to lost input. These responses were examined for evidence of disinhibition, a proposed mechanism of short-term visual plasticity. Border-zone probes elicited responses from regions of space immediately surrounding an artificial scotoma. These surrounding representations are proposed to provide input to scotoma representations. Border-zone responses were examined for correlates of invading activity from surrounding representations.

Perceptual filling-in often accompanies the conditioning of an artificial scotoma, and one might logically expect to find a reduced neural response from within its borders. However, disinhibition makes the opposite and somewhat counterintuitive prediction that visual responses will increase. Disinhibition induces increased neural response and enlarged receptive fields (Pettet and Gilbert 1992; DeAngelis et al. 1995). Thus, regardless of any perceptual filling-in that may occur, neurophysiological mechanisms predict an increased neural response. In accordance with this prediction, the amplitude of a visual sensory component, the P1, was enhanced relative to a stimulus-matched control condition. We propose that this P1 effect is the result of disinhibition within cortical scotoma-zone representations.

Though scotoma-zone visual responses increased, perceptual discrimination within scotoma boundaries decreased. This finding is consistent with previous psychophysical results (Kapadia et al. 1994; Mihaylov et al. 2007) but is seemingly contradictory to what would intuitively be expected. Enhanced visual responses (e.g., P1) are typically associated with improved perceptual performance (e.g., Luck et al. 1994). However, in the case of disinhibition, increased neural gain may come at the expense of response selectivity. Though there has not been a systematic investigation response tuning as a result of conditioning with an artificial scotoma, pharmacological blockade of local inhibitory inputs has been shown to increase neural response gain but decrease response selectivity (Sillito 1975). It may be possible that endogenous short-term disinhibition induces a similar state of nonselective gain. Mihaylov et al. (2007) previously demonstrated a reduction in the ability to discriminate a stimulus within an artificial scotoma, a result they attributed to an increased level of “internal noise” within scotoma-zone representations. Though the concept of internal noise is abstract in nature, the proposal is consistent with the expected impact of a loss of selectivity through disinhibition. Further research is necessary to fully understand the relationship between electrophysiological and psychophysical findings.

Previous neurophysiological studies of artificial scotomas in humans have demonstrated reductions in visual activity associated with the perceptual filling-in of the scotoma area (Weil et al. 2007, 2008). Weil et al. (2007) modulated scotoma luminance at 7.5 Hz and used steady state magnetoencephalography (MEG) responses to track the power of cortical frequency-domain signals during perceptual filling-in. They found a reduction in frequency-tagged signals during intervals where subjects reported filling-in of the artificial scotoma. Using the same paradigm, Weil et al. (2008) used fMRI to further localize visual areas correlated with the report of filling-in and found decreases in blood-oxygenation level dependent (BOLD) signals in early visual areas (V1/V2). In contrast to these findings, we demonstrate an increased visual response from within the boundaries of the scotoma zone, and the results of Weil et al. (2007, 2008) are seemingly at odds with our own. However, several key differences between their studies and ours make them complementary rather than contradictory. First, the theoretical questions being addressed are quite different. They report neural correlates of the phenomenological experience of the illusion of filling-in, whereas we were concerned with the short-term cortical dynamics, irrespective of the subjective experience of observers. Their results are specific to periods where filling-in is observed, whereas ours are related to the general functional changes associated with the conditioning of an artificial scotoma. Second, Weil et al. used a paradigm in which the scotoma region was contrast reversed. The neural signals evoked by this contrast reversal were likely driven at least in part by the edges of the scotoma region against the background. As such, reduced visual responses could be the result of adaptation of the scotoma edges rather than reflecting changes in scotoma-zone neural response properties. Lastly, it is difficult to directly compare the findings of Weil et al. with our own as there are considerable differences between the measures employed. They use steady state MEG potentials and BOLD signals to evaluate perceptual correlates developing over a period of scotoma conditioning, whereas we used ERP to probe changes in neural response properties following a constant conditioning period.

In Experiment 2, we sought a correlate of invading activity by probing the border zone of an artificial scotoma. Border-zone representations are expected to be a source of input into scotoma-zone representations. We reasoned that if inputs unmasked by disinhibition originated locally, rather than through feedback, then border-zone probes should exhibit differences within the same visual component as scotoma-zone probes (i.e., P1). However, if invading activity originates at a later stage of visual processing then border-zone probes should show effects in later visual responses (i.e., N1). Our results are in agreement with the latter hypothesis. ERPs derived from border-zone probes did not differ in P1 amplitude but exhibited a significant modulation in the later N1 component. We propose that this N1 effect serves as an index of invading activity and is indicative of feedback from a later stage of visual processing.

Local horizontal connections play a clear role in mediating topographical reorganization over the long term (Calford et al. 2003). Feedback has been proposed as another potential source of input driving long-term reorganization (Chino 1995; Darian-Smith and Gilbert 1995; Eysel and Schweigart 1999). Though these 2 proposals are not mutually exclusive, the role of feedback in long-term reorganization remains unclear, and local horizontal inputs appear to best account for changes in topographical representations in cats and monkeys (Chino 1995; Darian-Smith and Gilbert 1995; Young et al. 2002). It should be noted that, in the human visual system, feedback connections may play a more pronounced role in topographical reorganization (Masuda et al. 2008). Furthermore, in regards to short-term plasticity, neither the role of horizontal connections nor the feedback has been systematically investigated in any mammalian species (humans included). Single-cell studies of artificial scotomas generally interpret effects of plasticity arise via local horizontal connections, but the contribution of feedback versus local intrinsic connections has not been explored. Our ERP results suggest that feedback from later stages of visual processing may be an important component of short-term plasticity in the human visual system. Receptive field sizes increase with each subsequent stage in the human visual hierarchy (Smith et al. 2001). As such, later visual areas are well situated to provide early areas with input from broad regions of space. We do not exclude the involvement horizontal connections but simply propose that feedback from later stages of visual processing plays an important role in short-term plasticity. Additional work is necessary to further characterize the functional significance of such feedback and how that feedback relates to long-term cortical reorganization. We hypothesize that feedback may be particularly important in driving visual activity during an early phase of visual deafferentation and could potentially mediate, or guide, long-term topographical reorganization that occurs through other mechanisms (e.g., intrinsic horizontal connections).

ERP results demonstrated correlates of disinhibition within the scotoma zone (as indexed by P1) and correlates of invading activity arising from the border zone (indexed by N1). The occurrence and timing of these effects suggest the involvement of feedback from later stages of visual processing. However, ERP does not provide the spatial resolution necessary to accurately delineate the visual areas contributing to these effects. The cortical sources of P1 are thought to largely reflect activity from dorsal and ventral occipital temporal areas and those of N1 at least partially from similar sources but with additional parietal contributions as well (Di Russo et al. 2001). Thus, our results provide only crude spatial information and cannot pinpoint with accuracy from where in the visual hierarchy effects of short-term plasticity have occurred (e.g., V2, VP, V3A, V4). To accurately localize the cortical locus of each effect, a technique with much better spatial resolution must be employed (i.e., fMRI, MEG, or noninvasive optical imaging). Future research implementing such approaches is necessary to precisely determine the visual areas subserving short-term plasticity in the human brain.

In conclusion, we report 2 important findings of short-term plasticity in the human visual system using an artificial scotoma as a model of retinal deafferentation. First, ERPs revealed an increased visual response (indexed by P1) from within the boundaries of an artificial scotoma. Such a result is seemingly paradoxical to the “filling-in” percept that accompanies an artificial scotoma but is consistent with a proposed mechanism of short-term plasticity disinhibition. Second, regions of space surrounding an artificial scotoma (the border zone) showed an amplitude enhancement in a later component of the ERP—the N1. Border-zone representations provide invading input to visually deprived scotoma-zone representations. The occurrence and timing of this border-zone N1 effect suggest that unmasked activity of short-term plasticity arises, at least in part, from later stages of visual processing.

Funding

Dissertation Award granted by the American Psychological Association.

Conflict of Interest : None declared.

References

Angelucci
A
Bullier
J
Reaching beyond the classical receptive field of V1 neurons: horizontal and feedback axons?
J Physiol Paris
 , 
2003
, vol. 
97
 (pg. 
141
-
154
)
Arckens
L
Schweigart
G
Qu
Y
Wouters
G
Pow
DV
Vandesande
F
Eysel
UT
Orban
GA
Cooperative changes in GABA, glutamate and activity levels: the missing link in cortical plasticity
Eur J Neurosci
 , 
2000
, vol. 
12
 (pg. 
4222
-
4232
)
Baker
CI
Dilks
DD
Peli
E
Kanwisher
N
Reorganization of visual processing in macular degeneration: replication and clues about the role of foveal loss
Vision Res
 , 
2008
, vol. 
48
 (pg. 
1910
-
1919
)
Baker
CI
Peli
E
Knouf
N
Kanwisher
NG
Reorganization of visual processing in macular degeneration
J Neurosci
 , 
2005
, vol. 
25
 (pg. 
614
-
618
)
Calford
MB
Dynamic representational plasticity in sensory cortex
Neuroscience
 , 
2002
, vol. 
111
 (pg. 
709
-
738
)
Calford
MB
Schmid
LM
Rosa
MGP
Monocular focal retinal lesions induce short-term topographic plasticity in adult visual cortex
Proc Biol Sci
 , 
1999
, vol. 
266
 (pg. 
499
-
507
)
Calford
MB
Wang
C
Taglianetti
V
Waleszczyk
WJ
Burke
W
Dreher
B
Plasticity in adult cat visual cortex (area 17) following circumscribed monocular lesions of all retinal layers
J Physiol
 , 
2000
, vol. 
524
 (pg. 
587
-
602
)
Calford
MB
Wright
LL
Metha
AB
Taglianetti
V
Topographic plasticity in primary visual cortex is mediated by local corticocortical connections
J Neurosci
 , 
2003
, vol. 
23
 (pg. 
6434
-
6442
)
Chino
YM
Adult plasticity in the visual system
Can J Physiol Pharmacol
 , 
1995
, vol. 
73
 (pg. 
1323
-
1338
)
Chino
YM
Kaas
JH
Smith
EL
Langston
AL
Cheng
H
Rapid reorganization of cortical maps in adult cats following restricted deafferentation in retina
Vision Res
 , 
1992
, vol. 
32
 (pg. 
789
-
796
)
Chino
YM
Smith
EL
Kaas
JH
Sasaki
Y
Cheng
H
Receptive-field properties of deafferentated visual cortical neurons after topographic map reorganization in adult cats
J Neurosci
 , 
1995
, vol. 
15
 (pg. 
2417
-
2433
)
Darian-Smith
C
Gilbert
CD
Axonal sprouting accompanies functional reorganization in adult cat striate cortex
Nature
 , 
1994
, vol. 
368
 (pg. 
737
-
740
)
Darian-Smith
C
Gilbert
CD
Topographic reorganization in the striate cortex of the adult cat and monkey is cortically mediated
J Neurosci
 , 
1995
, vol. 
15
 (pg. 
1631
-
1647
)
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
, vol. 
375
 (pg. 
780
-
784
)
Das
A
Gilbert
CD
Receptive field expansion in adult visual cortex is linked to dynamic changes in strength of cortical connections
J Neurophysiol
 , 
1995
, vol. 
74
 (pg. 
779
-
792
)
DeAngelis
GC
Akiyuki
A
Ohzawa
I
Freeman
RD
Receptive field structure in the visual cortex: does selective stimulation induce plasticity
Proc Natl Acad Sci U S A
 , 
1995
, vol. 
92
 (pg. 
9682
-
9686
)
De Weerd
P
Perceptual filling-in: more than the eye can see
Prog Brain Res
 , 
2006
, vol. 
154
 (pg. 
227
-
245
)
De Weerd
P
Desimone
R
Ungerleider
LG
Perceptual filling-in: a parametric study
Vision Res
 , 
1998
, vol. 
38
 (pg. 
2721
-
2734
)
De Weerd
P
Gattass
R
Desimone
R
Ungerleider
LG
Responses of cells in monkey visual cortex during perceptual filling-in of an artificial scotoma
Nature
 , 
1995
, vol. 
377
 (pg. 
731
-
734
)
Di Russo
F
Mantinez
A
Sereno
MI
Pitzalis
S
Hillyard
SA
Cortical sources of the early components of the visual evoked potential
Hum Brain Mapp
 , 
2001
, vol. 
15
 (pg. 
95
-
111
)
Dilks
DD
Baker
CI
Peli
E
Kanwisher
N
Reorganization of visual processing in macular degeneration is not specific to the “preferred retinal locus”
J Neurosci
 , 
2009
, vol. 
29
 (pg. 
2768
-
2773
)
Dilks
DD
Serences
JT
Rosenau
BJ
Yantis
S
McCloskey
M
Human adult cortical reorganization and consequent visual distortion
J Neurosci
 , 
2007
, vol. 
27
 (pg. 
9585
-
9594
)
Dreher
B
Burke
W
Calford
MB
Cortical plasticity revealed by circumscribed retinal lesions or artificial scotomas
Prog Brain Res
 , 
2001
, vol. 
134
 (pg. 
217
-
245
)
Eysel
UT
Schweigart
G
Increased receptive field size in the surround of chronic lesions in the adult cat visual cortex
Cereb Cortex
 , 
1999
, vol. 
9
 (pg. 
101
-
109
)
Eysel
UT
Schweigart
G
Mittmann
T
Eyding
D
Qu
Y
Vandesande
F
Orban
G
Arckens
L
Reorganization in the visual cortex after retinal and cortical damage
Restor Neurol Neurosci
 , 
1999
, vol. 
15
 (pg. 
153
-
164
)
Gilbert
CD
Adult cortical dynamics
Physiol Rev
 , 
1998
, vol. 
78
 (pg. 
467
-
485
)
Gilbert
CD
Wiesel
TN
Receptive field dynamics in adult primary visual cortex
Nature
 , 
1992
, vol. 
356
 (pg. 
150
-
152
)
Heinen
SJ
Skavenski
AA
Recovery of visual responses in foveal V1 neurons following bilateral foveal lesions in adult monkey
Exp Brain Res
 , 
1991
, vol. 
83
 (pg. 
670
-
674
)
Kaas
JH
Krubitzer
LA
Chino
YM
Langston
AL
Polley
EH
Blair
N
Reorganization of retinotopic cortical maps in adult mammals after lesions of the retina
Science
 , 
1990
, vol. 
248
 (pg. 
229
-
231
)
Kapadia
MK
Gilbert
CD
Westheimer
G
A quantitative measure for short-term cortical plasticity in human vision
J Neurosci
 , 
1994
, vol. 
14
 (pg. 
451
-
457
)
Luck
SJ
Hillyard
SA
Mouloua
M
Woldorff
MG
Clark
VP
Hawkins
HL
Effects of spatial cuing on luminance detectability: psychophysical and electrophysiological evidence for early selection
J Exp Psychol Hum Percept Perform
 , 
1994
, vol. 
20
 (pg. 
887
-
904
)
Masuda
Y
Dumoulin
SO
Nakadomari
S
Wandell
BA
V1 projection zone signals in human macular degeneration depend on task, not stimulus
Cereb Cortex
 , 
2008
, vol. 
18
 (pg. 
2483
-
2493
)
Mihaylov
P
Manahilov
V
Simpson
WA
Strang
NC
Induced internal noise in perceptual artificial scotomas created by surrounding dynamic noise
Vision Res
 , 
2007
, vol. 
47
 (pg. 
1479
-
1489
)
Pettet
MW
Gilbert
CD
Dynamic changes in receptive-field size in cat primary visual cortex
Proc Natl Acad Sci U S A
 , 
1992
, vol. 
89
 (pg. 
8366
-
8370
)
Ramachandran
VS
Gregory
RL
Perceptual filling in of artificially induced scotomas in human vision
Nature
 , 
1991
, vol. 
350
 (pg. 
699
-
702
)
Rosier
AM
Arckens
L
Demeulemeester
H
Orban
GA
Eysel
UT
Wu
YJ
Vandesande
F
Effect of sensory deaffertation on immunoreactivity of GABAergic cells and on GABA receptors in the adult cat visual cortex
J Comp Neurol
 , 
1995
, vol. 
359
 (pg. 
476
-
489
)
Schmid
LM
Rosa
MGP
Calford
MB
Retinal-detachment induces massive immediate reorganization in visual cortex
Neuroreport
 , 
1995
, vol. 
65
 (pg. 
1349
-
1353
)
Schmid
LM
Rosa
MGP
Calford
MB
Ambler
JS
Visuotopic reorganization in the primary visual cortex of adult cats following monocular and binocular retinal lesions
Cereb Cortex
 , 
1996
, vol. 
6
 (pg. 
388
-
405
)
Schumacher
EH
Jacko
JH
Primo
SA
Main
KL
Moloney
KP
Kinzel
EN
Ginn
J
Reorganization of visual processing is related to eccentric viewing in patients with macular degeneration
Restor Neurol Neurosci
 , 
2008
, vol. 
26
 (pg. 
391
-
402
)
Sillito
AM
The contribution of inhibitory mechanisms to the receptive field properties of neurons in the striate cortex of the cat
J Physiol
 , 
1975
, vol. 
250
 (pg. 
305
-
329
)
Smirnakis
SM
Brewer
AA
Schmid
MC
Tolias
AS
Schüz
A
Augath
M
Inhoffen
W
Wandell
BA
Logothetis
NK
Lack of long-term cortical reorganization after macaque retinal lesions
Nature
 , 
2005
, vol. 
435
 (pg. 
300
-
307
)
Smith
AT
Singh
KD
Williams
AL
Greenlee
MW
Estimating receptive field size from fMRI data in human striate and extrastriate visual cortex
Cereb Cortex
 , 
2001
, vol. 
11
 (pg. 
1182
-
1190
)
Sunness
JS
Liu
T
Yantis
S
Retinotopic mapping of the visual cortex using functional magnetic resonance imaging in a patient with central scotomas from atrophic macular degeneration
Opthalmaology
 , 
2004
, vol. 
111
 (pg. 
1595
-
1598
)
Tremere
LA
De Weerd
P
Pinaud
R
Pinaud
R
Tremere
LA
De Weerd
P
A unified theoretical framework for plasticity in visual circuitry
Plasticity in the visual system: from genes to circuits
 , 
2005
New York
Springer-Verlag
(pg. 
347
-
355
)
Tremere
LA
Pinaud
R
De Weerd
P
Pessoa
L
De Weerd
P
Contributions of inhibitory mechanisms to perceptual completion and cortical reorganization
Filling-in from perceptual completion to cortical reorganization
 , 
2003
New York
Oxford University Press
(pg. 
295
-
323
)
Volchan
E
Gilbert
CD
Interocular transfer of receptive field expansion in cat visual cortex
Vision Res
 , 
1995
, vol. 
35
 (pg. 
1
-
6
)
Waleszczyk
WJ
Wang
C
Young
JM
Burke
W
Calford
MB
Dreher
B
Laminar differences in plasticity in area 17 following retinal lesions in kittens or adult cats
Eur J Neurosci
 , 
2003
, vol. 
17
 (pg. 
2351
-
2368
)
Weil
RS
Kilner
JM
Haynes
JD
Rees
G
Neural correlates of perceptual filling-in of an artificial scotoma in humans
Proc Natl Acad Sci U S A
 , 
2007
, vol. 
104
 (pg. 
5211
-
5216
)
Weil
RS
Watkins
S
Rees
G
Neural correlates of perceptual completion of an artificial scotoma in human visual cortex measured using function MRI
Neuroimage
 , 
2008
, vol. 
42
 (pg. 
1519
-
1528
)
Young
JM
Waleszczyk
WJ
Burke
W
Calford
MB
Dreher
B
Topographic reorganization in area 18 of adult cats following circumscribed monocular retinal lesions in adolescence
J Physiol
 , 
2002
, vol. 
541
 (pg. 
601
-
612
)
Zur
D
Ullman
S
Filling-in of retinal scotomas
Vision Res
 , 
2003
, vol. 
43
 (pg. 
971
-
982
)