Abstract

Using electrophysiology and a classic fear conditioning paradigm, this work examined adaptive visuocortical changes in spatial frequency tuning in a sample of 50 undergraduate students. High-density EEG was recorded while participants viewed 400 total trials of individually presented Gabor patches of 10 different spatial frequencies. Patches were flickered to produce sweep steady-state visual evoked potentials (ssVEPs) at a temporal frequency of 13.33 Hz, with stimulus contrast ramping up from 0% to 41% Michelson over the course of each 2800-msec trial. During the final 200 trials, a selected range of Gabor stimuli (either the lowest or highest spatial frequencies, manipulated between participants) were paired with an aversive 90-dB white noise auditory stimulus. Changes in spatial frequency tuning from before to after conditioning for paired and unpaired gratings were evaluated at the behavioral and electrophysiological level. Specifically, ssVEP amplitude changes were evaluated for lateral inhibition and generalization trends, whereas change in alpha band (8–12 Hz) activity was tested for a generalization trend across spatial frequencies, using permutation-controlled F contrasts. Overall time courses of the sweep ssVEP amplitude envelope and alpha-band power were orthogonal, and ssVEPs proved insensitive to spatial frequency conditioning. Alpha reduction (blocking) was most pronounced when viewing fear-conditioned spatial frequencies, with blocking decreasing along the gradient of spatial frequencies preceding conditioned frequencies, indicating generalization across spatial frequencies. Results suggest that alpha power reduction—conceptually linked to engagement of attention and alertness/arousal mechanisms—to fear-conditioned stimuli operates independently of low-level spatial frequency processing (indexed by ssVEPs) in primary visual cortex.

INTRODUCTION

Faced with an overwhelming flood of information, the human visual system must parse behaviorally relevant inputs from the constant sensory stream, granting selected items privileged access to motor and memory systems. Behavioral relevance may be driven by intrinsic stimulus properties (Zhang & Luck, 2009; Maunsell & Treue, 2006; Saenz, Buracas, & Boynton, 2002) or can be manipulated by task instructions, as is often done in studies of selective attention (e.g., Pourtois, Schettino, & Vuilleumier, 2013; Brosch, Pourtois, Sander, & Vuilleumier, 2011; Vuilleumier & Driver, 2007; Desimone & Duncan, 1995). Alternatively, stimuli may be prioritized based on learned behavioral relevance when they are reliably associated with threat or reward (Markovic, Anderson, & Todd, 2014; Miskovic & Keil, 2012; Hickey, Chelazzi, & Theeuwes, 2010). Examining the mechanisms underlying task-, feature-, and experience-based (learned) selection for visuocortical facilitation, this study measures neural population-level responses to spatial frequency-manipulated stimuli made motivationally relevant through classical conditioning (see Methods), with specific spatial frequencies predicting an aversive outcome. Spatial frequency sensitivity was also quantified behaviorally, both before and after conditioning, by means of the CSF (De Valois & De Valois, 1990; Campbell & Robson, 1968).

Previous work using orientation as the critical manipulated feature has demonstrated that aversively paired stimuli elicit heightened early visuocortical responses in both humans (Thigpen, Bartsch, & Keil, 2017; McTeague, Gruss, & Keil, 2015; Song & Keil, 2014) and monkeys (Li, Yan, Guo, & Li, 2019). As initial cortical visual processing is performed by overlapping functional channels selectively tuned to a range of spatial frequencies as well as orientations (Palmer, 1999; De Valois & De Valois, 1990; Blakemore & Campbell, 1969), here we tested whether aversively conditioning spatial frequency stimuli would mirror the behavioral and neural changes observed when orientation is the conditioned feature.

Spatial frequency, the level of stimulus detail per unit of visual angle (Blakemore & Campbell, 1969; Enroth-Cugell & Robson, 1966), is a fundamental stimulus feature necessary for optical processing that is both fast and accurate (Kauffmann, Chauvin, Pichat, & Peyrin, 2015; Mermillod, Guyader, & Chauvin, 2005; Schyns & Oliva, 1994; Watt, 1987). Low spatial frequency (LSF) information contains global, gist-like information, whereas high spatial frequency (HSF) content is necessary for detailed, fine-grained processing of images (Bar, 2007).

In addition to objective, stimulus-driven (exogenous, bottom–up) features such as spatial frequency, the allocation of limited neural processing bandwidth is also governed by goal-driven (endogenous, top–down) and contextual factors (Vuilleumier & Driver, 2007; Serences et al., 2005; Hopfinger, Buonocore, & Mangun, 2000; Egeth & Yantis, 1997; Theeuwes, 1994). Mounting evidence (e.g., Hintze, Junghöfer, & Bruchmann, 2014; Foxe & Simpson, 2002; Bullier, 2001; Kawasaki et al., 2001; Lamme & Roelfsema, 2000) of higher cortical activation (e.g., in pFC) occurring earlier in time than previously assumed possible underscores the importance of elucidating the role of higher order mechanisms in facilitated sensory processing. Here, we assess alpha band activity as an indicator of such higher order biasing processes along with steady-state visual evoked potential (ssVEP; see below) power as an index of early visual processing activity.

As the most ubiquitous oscillatory signal observed in the adult human EEG (Sadaghiani & Kleinschmidt, 2016), alpha oscillations are functionally correlated with sensory, motor, and memory processes (Başar & Güntekin, 2012; Klimesch, 2012) but have, from the earliest EEG studies (e.g., Adrian & Matthews, 1934), most frequently been linked to the facilitated processing of external stimuli. Alpha-band activity in the EEG may serve inhibitory functions (Haegens, Nácher, Luna, Romo, & Jensen, 2011; Jensen & Mazaheri, 2010), with decreased alpha power (often referred to as alpha blocking, alpha suppression or alpha desynchronization) generally observed in brain regions involved in task-related stimulus processing and/or regions where neuronal activity increases in response to task demands (Jensen, Bonnefond, & VanRullen, 2012; Klimesch, 2012; Jensen & Mazaheri, 2010). Increased alpha-band activity, on the other hand, is thought to reflect suppression of task-irrelevant and/or distracting information (Capilla, Schoffelen, Paterson, Thut, & Gross, 2012; Rihs, Michel, & Thut, 2007; Pfurtscheller & Da Silva, 1999). In direct contrast to the effects of ongoing, endogenous alpha-band activity just mentioned, exogenously induced alpha activity (as when driving ssVEPs within the alpha range) increases in amplitude along with spatial attention (Keitel et al., 2019). To preserve spontaneously occurring alpha activity, here we employed a ssVEP driving frequency of 13.33 Hz that lies outside the traditional alpha-band.

An oscillatory response to a periodically presented stimulus that adopts the same frequency as the driving stimulus (Regan, 1989), the main neural generators of the ssVEP (Figure 1) include striate and extrastriate visual cortical areas (Wieser & Keil, 2011; Di Russo et al., 2007,; Müller, Teder, & Hillyard, 1997) and exhibit significantly increased amplitudes both when attending to a driving (flickering) stimulus (Hajcak, MacNamara, Foti, Ferri, & Keil, 2013; Andersen, Hillyard, & Müller, 2008; Kim, Grabowecky, Paller, Muthu, & Suzuki, 2007; Müller, Picton, et al., 1998) and when the driving stimulus is emotionally arousing (Hajcak et al., 2013; Müller, Andersen, & Keil, 2007; Keil, Moratti, Sabatinelli, Bradley, & Lang, 2005). Importantly, the narrow-band ssVEP signal does not eliminate intrinsic brain oscillations, allowing assessment of visuocortical activity at the driven ssVEP frequency while simultaneously tracking ongoing alpha-band activity. Differential classical conditioning designs similar to that employed in this study have shown increased ssVEP amplitudes for oriented gratings paired with aversive (CS+) stimuli relative to unpaired (CS−) stimuli (e.g., McTeague et al., 2015; Hintze et al., 2014; Song & Keil, 2014), taken to indicate early cortical facilitation of the newly salient CS+ (reviewed in Miskovic & Keil, 2012). One hypothetical mechanism underlying potentiated CS+ amplitude responses in early visual cortex is reentrant signaling from anterior structures, perhaps originating from higher order cortical processes (Müller, Teder-Sälejärvi, & Hillyard, 1998) and/or arising through subcortical routes (Pessoa & Adolphs, 2010). Another, nonmutually exclusive possibility is that experience-mediated neuronal retuning in early visual cortex over time enhances responsiveness to the recurrent, distinctive features of the CS+ through mechanisms of short-term plasticity (Miskovic & Keil, 2012).

Figure 1. 

Extraction of ssVEP amplitude envelopes from time domain averages (band-pass filtered for illustration), using the filter Hilbert method. An example waveform is shown to illustrate the relation between the driven oscillations in the time domain and their time-varying envelope.

Figure 1. 

Extraction of ssVEP amplitude envelopes from time domain averages (band-pass filtered for illustration), using the filter Hilbert method. An example waveform is shown to illustrate the relation between the driven oscillations in the time domain and their time-varying envelope.

Investigating neural population-level orientation response properties following aversive conditioning, McTeague et al. (2015) found changes across a graded continuum of orientations surrounding the aversively paired CS+ to be consistent with a sharpened, difference-of-Gaussian (Mexican hat) tuning function. By contrast, when quantifying parietal ssVEP amplitudes as a proxy of attention system engagement, McTeague et al. observed a pattern of generalization rather than sharpening, with ssVEP amplitude responses to the CS+ orientation amplified and amplitudes gradually decreasing along with decreasing orientation similarity. A separate investigation (Rhodes, Ruiz, Ríos, Nguyen, & Miskovic, 2018) revealed that aversively conditioning grating orientations improves discrimination thresholds for orientations paired with an aversive noise compared with nonpaired orientations.

Following directly upon the findings outlined above, this study employed a paradigm with spatial frequency, rather than orientation, as the critical manipulated feature. Specifically, we tested the hypothesis that ssVEP envelopes evoked by a subset of flickering Gabors consistently paired with a noxious noise would show sharpened tuning properties toward the conditioned spatial frequencies. On the other hand, alpha power reduction (alpha blocking)—conceptually linked to facilitated cortical engagement, arousal, and attention—was expected to show generalization, with the greatest power reduction in response to CS+ gratings and blocking decreasing linearly with each log linear spatial frequency step away from conditioned gratings. Behaviorally, lowered detection contrast thresholds (measured as shorter RTs) were expected for conditioned relative to nonconditioned spatial frequency Gabors.

METHODS

Participants

Fifty-five undergraduate students (MAGE = 18.69 years, SDAGE = 0.91 years, 36 women) at the University of Florida participated in the study for course credit. To be included in the study, a participant had to have at least 25% of EEG trials (5/20) free of artifact (defined below) in each of the 10 spatial frequency conditions. Based on this criterion, data from five participants (all female) were not included in this sample, leaving a final sample of 50 (MAGE = 18.74 years, SDAGE = 0.92 years, 31 women) for EEG and subjective experience rating measures. Data from four additional participants (two women) are not included in contrast sensitivity analyses because of computer failure, leaving a final sample of 46 (MAGE = 18.76 years, SDAGE = 0.92 years, 29 women) for this measure. All participants reported normal or corrected-to-normal vision and no personal or family history of epilepsy or photic seizures. All participants provided informed consent in accordance with the Declaration of Helsinki and the institutional review board of the University of Florida.

Materials

Visual stimuli consisted of 10 (0.42, 0.61, 0.88, 1.27, 1.84, 2.66, 3.85, 5.58, 8.08, and 11.69 cycles/degree) gray-shaded, isoluminant, square wave Gabor patches presented against a gray background of the same mean luminance. Gabor/background contrast ramped up from 0% to 41% Michelson (preferred for contrast measures of gratings; Pelli & Farell, 1999) in log linear steps every 75 msec (i.e., with each presentation of the Gabor grating; Figure 2). Each Gabor was presented centrally, subtended a visual angle of 5.35° both horizontally and vertically, and was oriented at 45°. Stimuli were presented on a 23-in. 3-D LED monitor (Samsung LS23A950) set to a vertical refresh rate of 120 Hz connected to a PC running Windows XP. The unconditioned stimulus (US) consisted of a 90-dB SPL white noise delivered binaurally through two standard computer speakers located behind participants' seated location. Participants rated a subset of six Gabor patches on the dimensions of hedonic valence and emotional arousal using the self-assessment manikin (SAM; Bradley & Lang, 1994) 9-point scale both prior to and following the conditioning session.

Figure 2. 

Experimental procedure. Top, contrast sensitivity assessment; bottom left, example of one LSF habituation trial; bottom right, example of one LSF acquisition trial for participants in the LSF conditioning group.

Figure 2. 

Experimental procedure. Top, contrast sensitivity assessment; bottom left, example of one LSF habituation trial; bottom right, example of one LSF acquisition trial for participants in the LSF conditioning group.

Procedure

Participants were seated in a comfortable chair in a dimly lit (∼60 cd/m2) room with the viewing distance to the computer monitor adjusted to 130 cm. Participants completed the following tasks, in this order: (a) SAM rating six Gabor patches of different spatial frequencies, (b) initial contrast sensitivity assessment, (c) 200 trials of contrast sensitivity habituation, (d) 200 trials of contrast sensitivity acquisition/conditioning, (e) follow-up SAM ratings of the same set of Gabor stimuli used in (a), and (f) postconditioning contrast sensitivity assessment.

Contrast Sensitivity Assessment

Perception of spatial frequency information was assessed by finding detection threshold contrast levels over a range of Gabor-versus-background contrast levels for 10 spatial frequencies. Plotting these threshold values produces a U-shaped function, and taking their reciprocal yields the inverted-U-shaped spatial frequency CSF.

Participants were instructed that all Gabor stimuli would be presented centrally on-screen and that they should respond as quickly as possible upon detecting a pattern emerging from the uniform gray background (Figure 2, top). Detection was reported via left button click on a standard computer mouse. Within each block (pre- and postconditioning), participants completed three trials for each of the 10 spatial frequency conditions presented in pseudorandom order. EEG was not recorded during contrast sensitivity assessment.

Spatial Frequency Conditioning

In differential fear conditioning, an initially neutral stimulus acquires motivational relevance (becomes a conditioned stimulus, CS+) by being repeatedly paired with a naturally aversive stimulus (the US), whereas stimuli not associated with the US (CS−) remain neutral (reviewed in Miskovic & Keil, 2012).

Participants were instructed to remain as still as possible to limit blinking to the intertrial interval and were discouraged from making head or jaw movements during the experiment.

Spatial frequency conditioning comprised two blocks (Figure 2, bottom), each having 200 presentations of stimuli. The 10 spatial frequency conditions were presented in pseudorandom order, with each condition repeated 20 times in both habituation and acquisition blocks. Habituation trials made up the first block, followed by the acquisition trial block. The 10 centrally presented Gabor stimuli were flickered at a rate of 13.33 Hz and increased in contrast in log linear steps every 75 msec from 0% to 41% Michelson during each trial, producing sweep ssVEPs (e.g., Hamer & Norcia, 2009) at the driven oscillatory response frequency. Throughout the conditioning portion of the experiment, a small fixation spot remained in the center of the screen. The US was not presented in the habituation block. For the acquisition block, the US was paired with 100% of trials containing the two lowest (three highest) spatial frequency Gabor patches for the final 675 msec of such trials, with the US and the Gabor (CS+) coterminating at trial end. More densely sampling the high spatial frequency (SF) range with CS+ stimuli ensured that all participants perceived HSF gratings, thus addressing concerns based on the frequently reported observations that (i) the CSF drops off more sharply in the high SF, compared with the low SF range (e.g., Beazley, Illingworth, Jahn, & Greer, 1980) and (ii) there is more interindividual variability regarding sensitivity in the high SF than the low SF range (Ward, Rothen, Chang, & Kanai, 2017). All nonconditioned spatial frequency Gabor stimuli presentations proceeded unchanged from the habituation block.

EEG recording and preprocessing.

The EEG was recorded at a sampling rate of 500 Hz using a 128-channel Geodesic Sensor Net (EGI) system, with the vertex electrode (Cz) as recording reference. Impedances were kept below 50 kΩ as recommended by the manufacturer (Electrical Geodesics), and online data recording was constrained by Butterworth high-pass (3-dB point at 170 Hz) and low-pass (3-dB point at 0.05 Hz) filters.

Off-line, Butterworth low-pass (40 Hz, 18th order, 3-dB point at 45 Hz) and high-pass (4 Hz, second order, 3-dB point at 18 Hz) filters were applied. After arithmetically transforming the data to the average reference, artifact rejection was performed according to the statistical correction of artifacts in dense array studies (SCADS) procedure suggested by Junghöfer, Elbert, Tucker, and Rockstroh (2000), which uses statistical parameters including the distribution of the mean, standard deviation, and gradient of the voltage amplitude to identify outlying channels, trials, and trial–channel pairs. Sensors identified as contaminated with artifacts by this approach were replaced by statistically weighted, spherical spline interpolated values. The maximum number of interpolated channels in a given trial was set to 20, visually inspected to ensure that the rejected sensors were not located within one region of the scalp. All off-line filtering and preprocessing were performed using the EMEGS (Peyk, De Cesarei, & Junghöfer, 2011) MATLAB toolbox.

Analysis

The focus of analysis was on the overall pattern of electrophysiological and behavioral responses across the gradient of spatial frequencies approaching those frequencies designated as the CS+ in acquisition trials, rather than on the effect of LSF or HSF per se or on any pairwise comparisons between individual spatial frequencies or frequency ranges: In other words, conditioned SFs were varied as a counterbalancing measure. Thus, when investigating conditioning effects, all SF conditioning trials are arranged such that CS+ stimuli (either the lowest or highest SF Gabor patches) are the last conditions (the right flank of the distribution; see Figure 3), whereas the SFs farthest removed from the conditioned frequencies are first (the left flank). Exploratory analysis of ssVEP responses to individual spatial frequencies is also included in the Results section below. RTs maintain spatial frequency order to facilitate comparison with the traditional CSF. All outcome values reported, unless otherwise indicated, are change scores: EEG results are reported as the amplitude difference between acquisition and habituation trials (acquisition–habituation), whereas contrast sensitivity and subjective experience rating results are given as differences of values after conditioning minus values from a preconditioning assessment. Alpha-band activity is reported throughout as the power change from baseline after subtracting the average power in the 200 msec window preceding stimulus onset.

Figure 3. 

F-contrast weighting diagram. Left, lateral inhibition (aka Mexican hat) of ssVEP amplitude response; middle, generalization of ssVEP amplitude response; right, generalization of alpha-band amplitude response.

Figure 3. 

F-contrast weighting diagram. Left, lateral inhibition (aka Mexican hat) of ssVEP amplitude response; middle, generalization of ssVEP amplitude response; right, generalization of alpha-band amplitude response.

Contrast Sensitivity

Outlier trials were identified as those with RTs over 10 sec. After setting outlier RTs to 10 sec, the mean of each three-trial set of responses to each of the 10 spatial frequencies was used as each participant's RT for that particular spatial frequency. This procedure was used to obtain 10 mean RT change scores for each participant (post- minus preconditioning). Bayesian and frequentist 10 (SF Conditions) × 2 (Conditioning Group) repeated-measures ANOVAs were performed on RT change scores in JASP (JASP Team, 2018).

Subjective Experience Ratings

A generalization hypothesis was tested by fitting linear trend lines using the MATLAB function polyfit (with the SF Gabor farthest from the CS+ set as the intercept) to SAM valence and arousal rating change scores (post- minus preconditioning) across six rated Gabor patches ranging from most dissimilar to most similar to those serving as conditioned stimuli.

EEG

Trials for each of the 10 spatial frequencies were averaged across trials within participants, using data epochs of 2800 msec (200 msec pre- to 2600 msec poststimulus onset). Selection of topographical locations to include in analysis was accomplished through a spatial PCA. This PCA procedure was performed separately for ssVEP and alpha time series data using the average amplitude of the grand mean time course in a period from 2100 to 2700 msec poststimulus (the period of peak ssVEP amplitudes/maximum alpha reduction across all conditions). Thus, one 129-element (128 recorded channels plus the average reference) vector entered a 1000 (50 participants × 10 frequencies × 2 trial types) by 129 matrix, which was submitted to the default MATLAB function pca. Within participants, the first principal component of the omnibus PCA was taken as a sensor weighting vector and projected back onto the individual data time series, with the resulting 1-D time course measure representing the PCA-weighted average across the EEG sensors for each participant and condition.

ssVEP.

The amplitude envelope of the sweep ssVEP response targeted at 13.33 Hz was extracted via Hilbert transformation. A 10th order Butterworth filter was applied to the averaged data, and an analytic 90° phase-shifted signal of the band-pass filtered data was generated by means of the Hilbert transform (MATLAB function hilbert). The modulus of the original and phase-shifted data, computed for each time point, sensor, and spatial frequency condition, produced the time-varying ssVEP amplitude measure. Increasing the contrast of the Gabor gratings over time in each trial produces a sweep across contrast levels, with the resulting ssVEP envelope waveform serving as an estimate of the visuocortical contrast-response function at the neuronal population level (Hamer & Norcia, 2009).

Lateral inhibition or generalization of conditioning effects were separately tested for at each time point using F-contrasts (Rosnow, Rosenthal, & Rubin, 2000; Rosenthal & Rosnow, 1985), weighting each spatial frequency condition to account for the expected trend (Figure 3). Ranging from the SF most dissimilar to the CS+ to the lowest (highest) conditioned frequency, the weighting values were 0, −0.5, −0.5, −1, −2, −3, −2, 2, 3, and 4 for lateral inhibition, computed as a difference of Gaussians with an inner product equal to zero and −9, −7, −5, −3, −1, 1, 3, 5, 7, and 9 for generalization. Multiple comparisons were addressed by thresholding F values based on nonparametric permutation distributions for surrogate data (Maris & Oostenveld, 2007; Ernst, 2004), with spatial frequency conditions shuffled within participants. The 0.05 tail of a permutation distribution composed of 5000 shuffled F tests served as the significance threshold, with permutations performed using a well-established (e.g., McTeague et al., 2015; Heim & Keil, 2006) custom MATLAB script. Following trend analysis, the fit of lateral inhibition and generalization patterns across spatial frequencies was quantified with linear regression (MATLAB function fitlm).

Alpha-band power.

Alpha-band activity was estimated via Morlet wavelet convolution (Tallon-Baudry & Bertrand, 1999; Bertrand & Pantev, 1994). A Morlet constant (m; analysis frequency divided by the width of the wavelet in the frequency domain; f0f) of 7 was chosen to balance resolution of alpha activity in the time and frequency domains. Thus, wavelets with a center frequency between 7.93 and 12.07 Hz were used to quantify alpha-band changes. As the width of wavelets in the frequency domain changes as a function of m, frequency uncertainty was 0.57 Hz for the wavelet centered at 7.93 and 0.86 Hz for the wavelet centered at 12.07 Hz, whereas temporal uncertainty at these frequencies was 140 and 93 msec, respectively.

Planned F-contrast and regression analyses were conducted as with the ssVEP signal (detailed above), but testing only for a generalization trend of alpha blocking. Post hoc, quadratic and cubic regression models were also fit (including an intercept and all lower order terms for each predictor) to the alpha change data to probe the evident generalization trend.

RESULTS

Behavioral Results

Subjective Experience Ratings

Linear trends were first fit separately for each conditioning group. SAM valence ratings showed no difference in either intercept (p = .64, t(48) = 0.48, BF10 = 0.311) or slope (p = .99, t(48) = −0.017, BF10 = 0.28) between the two conditioning groups, so the groups were combined for the overall valence trend calculation. The intercept (i.e., Gabor most dissimilar to the CS+) value was 1.41, and both the intercept (p < .001, t(49) = 3.89, BF10 = 75.75) and slope of −0.76 (p < .001, t(49) = −5.02, BF10 = 2567.3) showed significant departure from zero in the expected directions. SAM arousal ratings from participants aversively conditioned to low SF Gabors displayed the expected trend, with an intercept of −4.27 and slope of 1.84 both differing significantly from zero (p < .001, t(22) = −6.58, BF10 = 14,142; p < .001, t(22) = 8.35, BF10 = 473,310). However, the overall trend for arousal ratings from participants in the high SF conditioning group remained flat, with neither the intercept (p = .63, t(26) = 0.49, BF10 = 0.23) nor the slope (p = .30, t(26) = 1.05, BF10 = 0.34) departing meaningfully from zero.

RTs

A repeated-measures ANOVA (Greenhouse–Geisser corrected for sphericity violations, denoted FGG, where appropriate) indicated—as expected, given the well-established properties of the CSF—a statistically significant effect of Spatial Frequency on RT change (p < .001, FGG(4.47, 210.11) = 5.54, MSE = 15.28, BF10 = 78,031.05; Figure 3). More importantly, however, there was no main effect of Conditioning Group (p = .31, F(1, 47) = 1.07, MSE = 5.19, BF10 = 0.26) nor was there a statistically significant interaction between Conditioning Group and Spatial Frequency (p = .66, FGG(4.47, 210.11) = 5.54, MSE = 1.73, BFinclusion = 0.01). Comparing RT improvement for each individual SF between conditioning groups with independent t tests revealed no differences (all uncorrected p values > .1 and BF10 values < 1), as did comparing RT change to conditioned versus nonconditioned stimuli collapsed across the two conditioning groups (p = .40, t(182) = −0.84, BF10 = 0.30; Figure 4, bottom inset).

Figure 4. 

Behavioral measures results. Top: SAM rating change scores by conditioning group. Bottom: Initial average RT for all participants (n = 46; black, solid line) versus postconditioning average RTs for participants in the low (n = 20; purple, dash-dotted line) or high (n = 26; yellow, dotted line) spatial frequency conditioning groups. Shaded area indicates the SEM. Bottom, inset: RT change averaged over conditioning groups. Bars show RT improvement for conditioned (0.42 and 0.61 c/degree for the low SF conditioned group, 8.08 and 11.69 c/degree for the high SF conditioned group) versus SF-matched nonconditioned Gabors. Error bars show the SEM.

Figure 4. 

Behavioral measures results. Top: SAM rating change scores by conditioning group. Bottom: Initial average RT for all participants (n = 46; black, solid line) versus postconditioning average RTs for participants in the low (n = 20; purple, dash-dotted line) or high (n = 26; yellow, dotted line) spatial frequency conditioning groups. Shaded area indicates the SEM. Bottom, inset: RT change averaged over conditioning groups. Bars show RT improvement for conditioned (0.42 and 0.61 c/degree for the low SF conditioned group, 8.08 and 11.69 c/degree for the high SF conditioned group) versus SF-matched nonconditioned Gabors. Error bars show the SEM.

EEG Results

Both the grand-mean amplitude envelope (Figure 5) and average time domain waveform (Figure 1) of the ssVEP tracked increasing stimulus contrast, whereas the average alpha response was unperturbed by higher stimulus contrast levels. The steady-state and alpha signals were also uncorrelated across time in every participant (Figure 6), never approaching significance thresholds determined by nonparametric permutation testing of Kendall's tau-b correlation values with SF conditions shuffled (n = 5000) within participants at all time points (see above).

Figure 5. 

Left: Sensor weightings for ssVEP and alpha, computed using the first principal component of a spatial PCA. Right: Grand mean PCA-weighted waveforms (n = 50). Shaded regions indicate the SEM. Only 100 msec of the 200 msec baseline period is shown here, as the first 100 msec contains the Hilbert filter onset artifact and is not representative of true ssVEP activity.

Figure 5. 

Left: Sensor weightings for ssVEP and alpha, computed using the first principal component of a spatial PCA. Right: Grand mean PCA-weighted waveforms (n = 50). Shaded regions indicate the SEM. Only 100 msec of the 200 msec baseline period is shown here, as the first 100 msec contains the Hilbert filter onset artifact and is not representative of true ssVEP activity.

Figure 6. 

Correlation between ssVEP and alpha amplitude for each participant across time over all conditions (arranged in order of increasing correlation value). Error bars show the SEM.

Figure 6. 

Correlation between ssVEP and alpha amplitude for each participant across time over all conditions (arranged in order of increasing correlation value). Error bars show the SEM.

ssVEP Amplitude Conditioning Effect

Permutation-controlled point-wise analyses confirmed that ssVEP amplitudes did not show a consistent learning-related change in spatial frequency tuning across all conditions at any time point (Figure 7, bottom). Converging with this result, neither the lateral inhibition nor the generalization model fit the ssVEP amplitude change values well (Table 1). Given the poor fit to the data, slope and intercept parameters are neither reported nor interpreted.

Figure 7. 

Top: Average 13.33 Hz ssVEP amplitude RT course for habituation (left) and acquisition (right) trials. For purposes of illustration, pairs of adjacent SF conditions were averaged. Bottom: Average 13.33 Hz ssVEP amplitude response change (acquisition minus habituation) over time (left) and over SF conditioning bins averaged over all time points (right). Error bars show the SEM.

Figure 7. 

Top: Average 13.33 Hz ssVEP amplitude RT course for habituation (left) and acquisition (right) trials. For purposes of illustration, pairs of adjacent SF conditions were averaged. Bottom: Average 13.33 Hz ssVEP amplitude response change (acquisition minus habituation) over time (left) and over SF conditioning bins averaged over all time points (right). Error bars show the SEM.

Table 1. 
13.33-Hz Steady-state Hilbert Envelope Average Model Fit Comparison
Time (msec)Deviance (Log Likelihood)BICF(1, 8)R2adj
Generalization model 
0:600 2.67 −0.73 3.04 0.16 
600:1200 4.51 −4.41 3.55 0.18 
1200:1800 −0.48 5.57 0.66 −0.05 
1800:2400 0.10 4.40 2.22 0.11 
  
Lateral inhibition model 
0:600 4.36 −4.11 3.47 0.19 
600:1200 5.95 −7.30 5.39 0.26 
1200:1800 1.20 2.20 0.99 −0.02 
1800:2400 1.90 0.80 3.42 0.17 
Time (msec)Deviance (Log Likelihood)BICF(1, 8)R2adj
Generalization model 
0:600 2.67 −0.73 3.04 0.16 
600:1200 4.51 −4.41 3.55 0.18 
1200:1800 −0.48 5.57 0.66 −0.05 
1800:2400 0.10 4.40 2.22 0.11 
  
Lateral inhibition model 
0:600 4.36 −4.11 3.47 0.19 
600:1200 5.95 −7.30 5.39 0.26 
1200:1800 1.20 2.20 0.99 −0.02 
1800:2400 1.90 0.80 3.42 0.17 

ssVEP Amplitude: Exploratory Analysis of Spatial Frequency Effects

Within a high-contrast window of 2200–2350 msec (chosen as the period of maximum amplitude response by inspection of the grand mean waveform), ssVEP amplitude responses across the range of SF stimuli varied little either as a function of conditioning block or conditioning group (Figure 8 and Table 2). One exception was at the highest SF (11.69 cpd), where the HSF conditioned group showed a markedly increased postconditioning amplitude response relative to the LSF conditioned group (p = .003, t(48) = 3.18, BF10 = 14.12).

Figure 8. 

Left: ssVEP amplitude response by spatial frequency for habituation (green squares) and acquisition (red stars) trial blocks. Error bars show the SEM. Right: ssVEP amplitude response by conditioning group. Initial (habituation block) average ssVEP amplitude response for all participants (n = 50; black, solid line) versus postconditioning ssVEP amplitude responses for participants in the low (n = 23; purple, dash-dotted line) or high (n = 27; yellow, dotted line) spatial frequency conditioning groups. Shaded area indicates the SEM.

Figure 8. 

Left: ssVEP amplitude response by spatial frequency for habituation (green squares) and acquisition (red stars) trial blocks. Error bars show the SEM. Right: ssVEP amplitude response by conditioning group. Initial (habituation block) average ssVEP amplitude response for all participants (n = 50; black, solid line) versus postconditioning ssVEP amplitude responses for participants in the low (n = 23; purple, dash-dotted line) or high (n = 27; yellow, dotted line) spatial frequency conditioning groups. Shaded area indicates the SEM.

Table 2. 
Steady-state Amplitude Change (Acq. − Hab.) by Spatial Frequency, 2200–2350 msec
SF (cpd)SDBF10t(48)p
0.42 0.76 0.31 0.49 .62 
0.61 0.68 0.37 −0.82 .42 
0.88 0.88 0.33 −0.58 .56 
1.27 1.06 0.35 0.71 .48 
1.84 0.87 0.41 0.95 .35 
2.66 0.83 0.28 −0.02 .98 
3.85 1.19 0.28 −0.11 .91 
5.58 1.33 0.28 −0.02 .99 
8.08 0.95 0.34 −0.68 .50 
11.69** 0.85 14.12 −3.18 .003 
SF (cpd)SDBF10t(48)p
0.42 0.76 0.31 0.49 .62 
0.61 0.68 0.37 −0.82 .42 
0.88 0.88 0.33 −0.58 .56 
1.27 1.06 0.35 0.71 .48 
1.84 0.87 0.41 0.95 .35 
2.66 0.83 0.28 −0.02 .98 
3.85 1.19 0.28 −0.11 .91 
5.58 1.33 0.28 −0.02 .99 
8.08 0.95 0.34 −0.68 .50 
11.69** 0.85 14.12 −3.18 .003 

Average inverse Bayes factor (BF10) and p values derived from independent-sample t tests comparing the LSF conditioned group to the HSF conditioned group.

**

Indicates statistical significance (p < .05) / strong support for the alternative hypothesis (BF10 > 10).

Alpha Power Conditioning Effect

F-contrast analysis showed that between 1888 and 2376 msec (Figure 9, lower left) the pointwise linear trend exceeded the permutation-derived critical F threshold value. Regression results indicated that the linear generalization model fit the data well (Figure 10 and Table 3).

Figure 9. 

Top: Average 7.93–12.07 Hz alpha-band amplitude RT course for habituation (left) and acquisition (right) trials. For purposes of illustration, pairs of adjacent SF conditions were averaged. Bottom: Average 7.93–12.07 Hz alpha-band amplitude response change (acquisition minus habituation) over time (left) and over SF conditioning bins averaged over all time points (right). Area between dotted lines (bottom, left) indicates time points where the linear trend of alpha change scores reached statistical significance as determined by permutation-controlled F-contrasts. Error bars (bottom, right) show the SEM.

Figure 9. 

Top: Average 7.93–12.07 Hz alpha-band amplitude RT course for habituation (left) and acquisition (right) trials. For purposes of illustration, pairs of adjacent SF conditions were averaged. Bottom: Average 7.93–12.07 Hz alpha-band amplitude response change (acquisition minus habituation) over time (left) and over SF conditioning bins averaged over all time points (right). Area between dotted lines (bottom, left) indicates time points where the linear trend of alpha change scores reached statistical significance as determined by permutation-controlled F-contrasts. Error bars (bottom, right) show the SEM.

Figure 10. 

Alpha-band amplitude change (acquisition minus habituation) linear generalization trend (dotted line) and average alpha amplitude change values (green dots) for indicated time intervals. Error bars show the SEM.

Figure 10. 

Alpha-band amplitude change (acquisition minus habituation) linear generalization trend (dotted line) and average alpha amplitude change values (green dots) for indicated time intervals. Error bars show the SEM.

Table 3. 
Alpha Generalization Model Fit Averages
Time (msec)Deviance (Log Likelihood)BICF(1, 8)R2adj
0:600 −18.06 40.72 5.45 0.24 
600:1200 −19.59 43.78 9.57 0.54 
1200:1800 −24.75 54.10 2.05 0.19 
1800:2400 −25.06 54.73 8.47 0.50 
  
Alpha Generalization Model, Average Slope, and Intercept 
Time (msec)ParameterEstimate (SE)log(BF10)p
0:600 Slope −0.12 (0.10) 125.11 <.001 
Intercept −0.06 (0.45) 1.48 .004 
600:1200 Slope −0.33 (0.11) 515.92 <.001 
Intercept 0.48 (0.50) 256.01 <.001 
1200:1800 Slope −0.24 (0.18) 401.20 <.001 
Intercept 0.05 (0.84) 0.17 .02 
1800:2400 Slope −0.52 (0.18) 666.10 <.001 
Intercept 1.81 (0.86) 529.18 <.001 
Time (msec)Deviance (Log Likelihood)BICF(1, 8)R2adj
0:600 −18.06 40.72 5.45 0.24 
600:1200 −19.59 43.78 9.57 0.54 
1200:1800 −24.75 54.10 2.05 0.19 
1800:2400 −25.06 54.73 8.47 0.50 
  
Alpha Generalization Model, Average Slope, and Intercept 
Time (msec)ParameterEstimate (SE)log(BF10)p
0:600 Slope −0.12 (0.10) 125.11 <.001 
Intercept −0.06 (0.45) 1.48 .004 
600:1200 Slope −0.33 (0.11) 515.92 <.001 
Intercept 0.48 (0.50) 256.01 <.001 
1200:1800 Slope −0.24 (0.18) 401.20 <.001 
Intercept 0.05 (0.84) 0.17 .02 
1800:2400 Slope −0.52 (0.18) 666.10 <.001 
Intercept 1.81 (0.86) 529.18 <.001 

Average inverse Bayes factor (BF10) and p values derived from one-sample t tests of intercept and slope values versus 0 (i.e., no change in power from habituation to acquisition trials).

Comparing linear trends for the two conditioning groups fit separately, alpha power change values showed no difference in either intercept (p = .56, t(49) = 0.14, BF10 = 0.29) or slope (p = .68, t(48) = 0.41, BF10 = 0.30), so the groups were combined into one overall generalization model. Parameter estimates from four averaged time windows (600 msec each) created post hoc (Table 3) all show negative slope values (all p ≤ .001 and log(BF10) ≥ 125.11) along with significant intercept values (all p ≤ .02 and log(BF10) ≥ 0.17). Higher order model fit measures are reported in Table 4, and the fit to the data over all time points for both the linear and the cubic models is shown in Figure 11.

Table 4. 
Alpha-band Power Change Model Comparison (Average over All Time Points and Participants)
TrendDeviance (Log Likelihood)BICAICF(df)R2adj
Linear −21.91 48.43 47.83 6.05 (1, 8) 0.27 
Quadratic −20.48 47.86 46.96 5.25 (2, 7) 0.49 
Cubic −18.06 45.33 44.12 6.92 (3, 6) 0.65 
TrendDeviance (Log Likelihood)BICAICF(df)R2adj
Linear −21.91 48.43 47.83 6.05 (1, 8) 0.27 
Quadratic −20.48 47.86 46.96 5.25 (2, 7) 0.49 
Cubic −18.06 45.33 44.12 6.92 (3, 6) 0.65 
Figure 11. 

Top: Alpha-band change (acquisition – habituation) goodness-of-fit comparison. Adjusted R2 values over all time points shown for cubic (gold, solid line) and linear (purple, dotted line) trends. Bottom: Comparison of cubic and linear fits over SF conditioning bins averaged over indicated time points.

Figure 11. 

Top: Alpha-band change (acquisition – habituation) goodness-of-fit comparison. Adjusted R2 values over all time points shown for cubic (gold, solid line) and linear (purple, dotted line) trends. Bottom: Comparison of cubic and linear fits over SF conditioning bins averaged over indicated time points.

DISCUSSION

This study set out to examine experience-related changes in spatial frequency tuning indexed by behavioral and electrophysiological measures. A single conditioning session prompted selective changes in alpha-band power reduction that was most pronounced when viewing conditioned spatial frequencies. By contrast, psychophysical testing of contrast sensitivity did not show selective changes, nor did the neuronal population-level visuocortical CSF measured by means of sweep ssVEPs. Importantly, sweep ssVEP envelopes across spatial frequencies showed a gradual increase as stimulus contrast ramped up over the course of a trial, whereas the time course of alpha reduction did not increase with contrast and was not related to the time-varying magnitude of the ssVEP signal. Thus, whereas ssVEPs reflect a contrast-dependent neural mechanism, alpha reduction during visual cue conditioning appears to reflect a higher order facilitation mechanism, which is sensitive to spatial frequency conditioning. In the following, we briefly discuss the implications of the findings for each dependent variable.

Subjective Experience Ratings

Overall, changes in self-reported valence and intensity ratings showed the expected trends, with participants reporting more intense and less pleasant responses towards stimuli resembling the CS+ (Figure 4, top). In terms of hedonic valence (pleasure), Gabor gratings most unlike the CS+ increased by an average of 1.41 points, with scores decreasing by approximately three fourths of a point for each SF step toward the CS+. Thus, both conditioning groups reported valence scores that were in line with experimental contingencies. This suggests that, as a group, participants learned that certain spatial frequencies were predictive of the noxious noise.

Contrast Sensitivity (RTs)

Plotting RTs across increasing spatial frequencies produced a canonical CSF (Figure 4, bottom), indicating that the range of SFs used here provide a suitable index of contrast sensitivity. Whereas improvements in psychophysical response functions are seen over multiple sessions of discrimination training for visual features such as orientation and motion (e.g., Adini, Sagi, & Tsodyks, 2002; Sagi & Tanne, 1994), the shape of the overall CSF was not modified within a single experimental session of aversive conditioning. Instead, the overall postconditioning CSF response gain evident in Figure 3 is consistent with previous accounts of global spatial frequency sensitivity improvements linked to covert attention (Cameron, Tai, & Carrasco, 2002; Carrasco, Penpeci-Talgar, & Eckstein, 2000), although in this study participants were free to utilize overt attention and foveate all centrally presented stimuli. The observation that aversive conditioning did not affect the shape of the CSF dovetails with ssVEP results discussed next, which also suggest that spatial frequency tuning properties in primary visual cortex are not altered within a single, brief conditioning session.

ssVEP Amplitude

Against expectations, ssVEP amplitude changes did not show a pattern consistent with either lateral inhibition or generalization. These results were surprising, as tuning properties for orientation and spatial frequency are known to be closely linked (Zhu, Xing, Shelley, & Shapley, 2010; Xing, Ringach, Shapley, & Hawken, 2004; De Valois, Albrecht, & Thorell, 1982) and ssVEP amplitude changes are sensitive to orientation conditioning.

Discrepant conditioning effects observed in orientation versus spatial frequency conditioning may be due to differing neural loci of tuning for the two properties. Whereas orientation selectivity is generated in V1 (Ringach, Hawken, & Shapley, 1997; De Valois et al., 1982; Hubel & Wiesel, 1962), the spatial frequency content of visual images is selectively filtered from the first moment of processing in retinal ganglion cells (Enroth-Cugell & Robson, 1966; Kuffler, 1953) and becomes more frequency selective (narrowly tuned) in the LGN (So & Shapley, 1981; Hubel & Wiesel, 1977; Blakemore & Campbell, 1969) before passing into primary visual cortex. Although some frequency tuning appears to continue at the cortical level, as neural channels in V1 exhibit more frequency selectivity than do single neurons in the LGN (De Valois & De Valois, 1990; Movshon, Thompson, & Tolhurst, 1978; Blakemore & Campbell, 1969), the degree of spatial frequency tuning achieved during the feed-forward sweep from retina to cortex versus how much is accomplished via intracortical mechanisms remains undetermined (Miller, 2003; Ringach, Bredfeldt, Shapley, & Hawken, 2002).

Additionally, differences in experimental protocol between the current study and that of orientation tuning done by McTeague et al. (2015) may explain the divergent outcomes with respect to ssVEP amplitudes. Most notably, McTeague et al. used higher contrast (95%) Gabor stimuli set against a black background, whereas here the maximum contrast reached 41% against a dark gray background. Suggesting that contrast differences alone cannot explain these differing outcomes, however, Song and Keil (2014) observed ssVEP enhancement of conditioned orientations using stimuli and maximum contrast levels nearly identical to those in the present investigation.

Interestingly, at high contrast levels the ssVEP response across the range of SF stimuli used here bears a remarkable similarity to the RT-derived CSF (Figure 4 and Figure 8, right). Also notable from our exploratory analyses, LSF conditioning failed to promote an enhanced early visual cortical response to LSF stimuli. LSF content is often held to preferentially convey emotional information (Burra, Hervais-Adelman, Celeghin, de Gelder, & Pegna, 2017; Jessen & Grossmann, 2017; Vuilleumier, Armony, Driver, & Dolan, 2003), and several influential models (e.g., LeDoux, 2000; Morris et al., 1998) stress the importance of amygdala activation via LSF-dominant magnocellular visual channels in amplified visual area responses to emotional objects. Both that LSF information preferentially elicits emotional responses (reviewed in De Cesarei & Codispoti, 2013) and that manipulating spatial frequency cleanly isolates the magno- and parvocellular visual streams (reviewed in Skottun, 2015) remain controversial claims, however. Results here fail to support any advantage for LSF over HSF visual information in modulating the neural response to aversively conditioned spatial frequency stimuli in primary visual cortex.

Alpha Power

As predicted, alpha blocking was selectively enhanced for CS+ stimuli. Furthermore, this conditioned alpha-band power reduction was found to gradually diminish across the gradient of Gabor patches, with decreasing similarity to the CS+. Results reported here accord with the established role of alpha blocking in amplifying sensory signals for behaviorally relevant objects and add to this literature base in several respects.

Alpha blocking has been shown to correlate with enhanced cortical processing of low-level stimulus properties such as color and motion when these dimensions are made task-relevant (Snyder & Foxe, 2010), and here we demonstrate that the alpha-band response is also sensitive to SF-associated contingencies. In addition, although alpha suppression in response to naturally foreboding stimuli is well established (e.g., Vagnoni, Lourenco, & Longo, 2015; De Cesarei & Codispoti, 2011; Huster, Stevens, Gerlach, & Rist, 2009; Güntekin & Basar, 2007), only recently (Panitz, Keil, & Mueller, 2019) has this effect been linked to conditioned threat cues. The present work shows that conditioned alpha blocking extends to a visual feature as basic as spatial frequency when such information provides the only basis for discriminating threatening from nonthreatening stimuli. Imbuing inherently neutral Gabor patches with a learned threat association allows us to conclude that the neural modulation observed in this study is driven by the conditioning experience, unconfounded by exogenous, stimulus-related factors.

Although a pronounced downward trend in alpha power is evident early within trials when contrast is low (Figure 10), at high stimulus contrast (later time points), the alpha power bifurcates around the middle of the spatial frequency conditions, with alpha power responses increased to distinctly CS− stimuli and decreased to CS+ and similar stimuli. This categorical pattern is consistent with an arousal-based perspective of attention and perception (reviewed in Mather & Sutherland, 2011). This notion posits that when stimuli do not compete for perceptual processing bandwidth, arousal enhances the processing of neutral but task-relevant stimuli while reducing processing of non-relevant stimuli (Mather & Sutherland, 2011). Exploratory higher order functions (quadratic and cubic) described the alpha-blocking generalization trend better than linear trends, even after accounting for added model complexity (Figure 11 and Table 4). The advantage of these higher order functions over a linear description emerges primarily at medium to high contrast, where a curvilinear shape is better able to capture both the step-like response change around the middle of the spatial frequency conditioning gradient, as well as the upturn occurring at the final, conditioned frequency (see Figure 11).

Relation between Alpha-band Power and ssVEP Amplitude Time Courses

Driving the ssVEP at a frequency above the traditional alpha band, the time course of the steady-state signal was uncorrelated with the alpha power response to increasing contrast (Figure 6). After decreasing for the first ∼600 msec following stimulus onset (Figure 9, top), alpha-band power changes displayed a step function-like response across frequency conditions (Figure 10) absent further overall power decreases. The ssVEP response, on the other hand, was positively related to contrast, with both increasing over the course of trials.

Conclusions

Exploring the development of experience-mediated visuocortical facilitation, the current work found alpha power to be sensitive to SF-associated aversive conditioning. Specifically, alpha blocking toward conditioned stimuli showed robust generalization, though—especially at higher background\stimulus contrast levels—the power change trend was better described by a cubic function. One possibility is that the step-like learning gradient captured by a cubic trend represents a learned process of categorical perception of cues as either safe or unsafe, a dichotomous distinction that is more apparent at high contrast. On the other hand, ssVEP amplitudes were not sensitive to SF conditioning, and the present conditioning paradigm did not prompt significant alteration in psychophysically assessed CSFs.

Results suggest a dissociation between visual tuning properties for orientation, which is altered by aversive conditioning, and spatial frequency. Furthermore, the fact that a high-order, contrast-independent mechanism indexed by alpha power reduction was sensitive to SF conditioning, whereas sweep ssVEPs were not, raises several pertinent questions for future consideration. For example, in the absence of visuocortical gain changes, selective alpha power reduction may reflect facilitation outside the main generators of the ssVEP (i.e., outside primary visual cortex), but it is also conceivable that primary visual areas contribute to alpha reduction in ways that are not captured by the sweep ssVEP method. Consistent with recent notions of alpha oscillations as carriers of interarea communication signals (Chapeton, Haque, Wittig, Inati, & Zaghloul, 2019), it is also plausible that the observed changes in alpha reduction reflect changes in frontoparietal connectivity, as previously observed during visual aversive conditioning (Petro et al., 2017).

Limitations and Directions for Future Studies

Given the primacy of both spatial frequency and orientation in psychophysically inspired accounts of early visual processing (Palmer, 1999; De Valois & De Valois, 1990; Blakemore & Campbell, 1969), further clarification of each of these properties' respective contribution to enhanced processing of motivationally relevant stimuli is desirable. One testable hypothesis following from the present work is that both spatial frequency and orientation conditioning similarly impact higher order neurophysiological processes but differentially influence lower levels of the hierarchy. Orientation conditioning primarily alters the first afferent volley into primary visual cortex (Li et al., 2019; Thigpen et al., 2017), supporting the notion that orientation selectivity arises at the macroscopic level of V1 population activity. By contrast, spatial frequency conditioning may alter neural responses at subcortical processing stages or selectively at the microscopic level and may require additional time and more trials to do so.

In the present experimental protocol, trial time and contrast were directly related and thus confounded. Future efforts to dissociate the effects of time and contrast on ssVEP amplitudes could manipulate maximum contrast to occur at different time points within trials. An investigation using static high-contrast stimuli would also be informative in determining whether ssVEP amplitude responses are sensitive to SF conditioning at higher contrasts.

As there was no predetermined time window of interest, alpha power trends shown in Figure 10 and reported in Table 3 were estimated on activity periods determined by post hoc inspection of the data. Although such a process could potentially increase false positive findings by “double-dipping” (Kriegeskorte, Simmons, Bellgowan, & Baker, 2009) into the data, the overall results and interpretation of alpha activity presented here hold regardless of the time window considered. Additionally, precise characterization of the higher level process(es) indexed by alpha blocking was not undertaken in the current work. Selective attention, alertness, and/or arousal are all credible candidates, but as these closely related processes are often coactive (Sadaghiani & Kleinschmidt, 2016), distinguishing between them often requires specific task manipulations in conjunction with neuroanatomical imaging and/or neurochemical assays (Sadaghiani & Kleinschmidt, 2016; Posner, 2008).

A further limitation arises from the fact there was no task during the EEG recording, where one might expect a systematic relation between alpha decrease and facilitated behavior to mainfest as faster RTs. We avoided a task during the ramping-up phase of the Gabor to eliminate potentially confounding detection- and response-related processes.

Analyses of averaged data necessarily trade information on trial-by-trial changes for statistical power. Previous single trial work (Liu, Keil, & Ding, 2012) suggests, however, that fear conditioning prompts a phasic amplitude change in ongoing activity detectable at the scalp, with early amplitude response gains to the CS+ followed by a subsequent return to near baseline levels. As amplitude enhancements for CS+ stimuli have been observed in steady-state responses following as few as 10 conditioning trial associations (Moratti & Keil, 2005; Moratti, Keil, & Miller, 2006), future efforts to further characterize the aversive conditioning process as it develops over time through single-trial analyses are warranted.

Acknowledgments

This work was supported by grants from the National Institutes of Health R01MH112558 and R01 MH097320 and grant N00014-18-1-2306 from the Office of Naval Research to A. K.

Reprint requests should be sent to Wendel M. Friedl, University of Florida, 3063 Long Leaf Rd., Building 772, Gainesville, FL 32608, or via e-mail: wfriedl@ufl.edu.

REFERENCES

Adini
,
Y.
,
Sagi
,
D.
, &
Tsodyks
,
M.
(
2002
).
Context-enabled learning in the human visual system
.
Nature
,
415
,
790
793
.
Adrian
,
E. D.
, &
Matthews
,
B. H.
(
1934
).
The Berger rhythm: Potential changes from the occipital lobes in man
.
Brain
,
57
,
355
385
.
Andersen
,
S. K.
,
Hillyard
,
S. A.
, &
Müller
,
M. M.
(
2008
).
Attention facilitates multiple stimulus features in parallel in human visual cortex
.
Current Biology
,
18
,
1006
1009
.
Bar
,
M.
(
2007
).
The proactive brain: Using analogies and associations to generate predictions
.
Trends in Cognitive Sciences
,
11
,
280
289
.
Başar
,
E.
, &
Güntekin
,
B.
(
2012
).
A short review of alpha activity in cognitive processes and in cognitive impairment
.
International Journal of Psychophysiology
,
86
,
25
38
.
Beazley
,
L. D.
,
Illingworth
,
D. J.
,
Jahn
,
A.
, &
Greer
,
D. V.
(
1980
).
Contrast sensitivity in children and adults
.
British Journal of Ophthalmology
,
64
,
863
866
.
Bertrand
,
O.
, &
Pantev
,
C.
(
1994
).
Stimulus frequency dependence of the transient oscillatory auditory evoked responses (40 Hz) studied by electric and magnetic recordings in human
. In
C.
Pantev
,
T.
Elbert
, &
B.
Lütkenhöner
(Eds.),
Oscillatory event-related brain dynamics
(
Vol. 271
, pp.
231
242
).
New York
:
Springer
.
Blakemore
,
C.
, &
Campbell
,
F. W.
(
1969
).
On the existence of neurones in the human visual system selectively sensitive to the orientation and size of retinal images
.
Journal of Physiology
,
203
,
237
260
.
Bradley
,
M. M.
, &
Lang
,
P. J.
(
1994
).
Measuring emotion: The self-assessment manikin and the semantic differential
.
Journal of Behavioral Therapy and Experimental Psychiatry
,
25
,
49
59
.
Brosch
,
T.
,
Pourtois
,
G.
,
Sander
,
D.
, &
Vuilleumier
,
P.
(
2011
).
Additive effects of emotional, endogenous, and exogenous attention: Behavioral and electrophysiological evidence
.
Neuropsychologia
,
49
,
1779
1787
.
Bullier
,
J.
(
2001
).
Integrated model of visual processing
.
Brain Research Reviews
,
36
,
96
107
.
Burra
,
N.
,
Hervais-Adelman
,
A.
,
Celeghin
,
A.
,
de Gelder
,
B.
, &
Pegna
,
A. J.
(
2017
).
Affective blindsight relies on low spatial frequencies
.
Neuropsychologia
,
128
,
44
49
.
Cameron
,
E. L.
,
Tai
,
J. C.
, &
Carrasco
,
M.
(
2002
).
Covert attention affects the psychometric function of contrast sensitivity
.
Vision Research
,
42
,
949
967
.
Campbell
,
F. W.
, &
Robson
,
J. G.
(
1968
).
Application of Fourier analysis to the visibility of gratings
.
Journal of Physiology
,
197
,
551
566
.
Capilla
,
A.
,
Schoffelen
,
J. M.
,
Paterson
,
G.
,
Thut
,
G.
, &
Gross
,
J.
(
2012
).
Dissociated α-band modulations in the dorsal and ventral visual pathways in visuospatial attention and perception
.
Cerebral Cortex
,
24
,
550
561
.
Carrasco
,
M.
,
Penpeci-Talgar
,
C.
, &
Eckstein
,
M.
(
2000
).
Spatial covert attention increases contrast sensitivity across the CSF: Support for signal enhancement
.
Vision Research
,
40
,
1203
1215
.
Chapeton
,
J. I.
,
Haque
,
R.
,
Wittig
,
J. H.
,
Inati
,
S. K.
, &
Zaghloul
,
K. A.
(
2019
).
Large-scale communication in the human brain is rhythmically modulated through alpha coherence
.
Current Biology
,
29
,
2801
2811
.
De Cesarei
,
A.
, &
Codispoti
,
M.
(
2011
).
Affective modulation of the LPP and α-ERD during picture viewing
.
Psychophysiology
,
48
,
1397
1404
.
De Cesarei
,
A.
, &
Codispoti
,
M.
(
2013
).
Spatial frequencies and emotional perception
.
Reviews in the Neurosciences
,
24
,
89
104
.
De Valois
,
R. L.
,
Albrecht
,
D. G.
, &
Thorell
,
L. G.
(
1982
).
Spatial frequency selectivity of cells in macaque visual cortex
.
Vision Research
,
22
,
545
559
.
De Valois
,
R. L.
, &
De Valois
,
K. K.
(
1990
).
Spatial vision
.
Oxford, United Kingdom
:
Oxford University Press
.
Desimone
,
R.
, &
Duncan
,
J.
(
1995
).
Neural mechanisms of selective visual attention
.
Annual Review of Neuroscience
,
18
,
193
222
.
Di Russo
,
F.
,
Pitzalis
,
S.
,
Aprile
,
T.
,
Spitoni
,
G.
,
Patria
,
F.
,
Stella
,
A.
, et al
(
2007
).
Spatiotemporal analysis of the cortical sources of the steady-state visual evoked potential
.
Human Brain Mapping
,
28
,
323
334
.
Egeth
,
H. E.
, &
Yantis
,
S.
(
1997
).
Visual attention: Control, representation, and time course
.
Annual Review of Psychology
,
48
,
269
297
.
Enroth-Cugell
,
C.
, &
Robson
,
J.
(
1966
).
The contrast sensitivity of retinal ganglion cells of the cat
.
Journal of Physiology
,
187
,
517
552
.
Ernst
,
M. D.
(
2004
).
Permutation methods: A basis for exact inference
.
Statistical Science
,
19
,
676
685
.
Foxe
,
J. J.
, &
Simpson
,
G. V.
(
2002
).
Flow of activation from V1 to frontal cortex in humans
.
Experimental Brain Research
,
142
,
139
150
.
Güntekin
,
B.
, &
Basar
,
E.
(
2007
).
Emotional face expressions are differentiated with brain oscillations
.
International Journal of Psychophysiology
,
64
,
91
100
.
Haegens
,
S.
,
Nácher
,
V.
,
Luna
,
R.
,
Romo
,
R.
, &
Jensen
,
O.
(
2011
).
α-Oscillations in the monkey sensorimotor network influence discrimination performance by rhythmical inhibition of neuronal spiking
.
Proceedings of the National Academy of Sciences, U.S.A.
,
108
,
19377
19382
.
Hajcak
,
G.
,
MacNamara
,
A.
,
Foti
,
D.
,
Ferri
,
J.
, &
Keil
,
A.
(
2013
).
The dynamic allocation of attention to emotion: Simultaneous and independent evidence from the late positive potential and steady state visual evoked potentials
.
Biological psychology
,
92
,
447
455
.
Hamer
,
R. D.
, &
Norcia
,
A. M.
(
2009
).
The Jitter Spatial Frequency Sweep VEP: A new paradigm to study spatiotemporal development of pattern- and motion-processing mechanisms in human infants
.
Psychology & Neuroscience
,
2
,
163
177
.
Heim
,
S.
, &
Keil
,
A.
(
2006
).
Effects of classical conditioning on identification and cortical processing of speech syllables
.
Experimental Brain Research
,
175
,
411
424
.
Hickey
,
C.
,
Chelazzi
,
L.
, &
Theeuwes
,
J.
(
2010
).
Reward changes salience in human vision via the anterior cingulate
.
Journal of Neuroscience
,
30
,
11096
11103
.
Hintze
,
P.
,
Junghöfer
,
M.
, &
Bruchmann
,
M.
(
2014
).
Evidence for rapid prefrontal emotional evaluation from visual evoked responses to conditioned gratings
.
Biological Psychology
,
99
,
125
136
.
Hopfinger
,
J. B.
,
Buonocore
,
M. H.
, &
Mangun
,
G. R.
(
2000
).
The neural mechanisms of top–down attentional control
.
Nature Neuroscience
,
3
,
284
291
.
Hubel
,
D. H.
, &
Wiesel
,
T. N.
(
1962
).
Receptive fields, binocular interaction and functional architecture in the cat's visual cortex
.
Journal of Physiology
,
160
,
106
154
.
Hubel
,
D. H.
, &
Wiesel
,
T. N.
(
1977
).
Ferrier lecture: Functional architecture of macaque monkey visual cortex
.
Proceedings of the Royal Society of London, Series B: Biological Sciences
,
98
,
1
59
.
Huster
,
R. J.
,
Stevens
,
S.
,
Gerlach
,
A. L.
, &
Rist
,
F.
(
2009
).
A spectralanalytic approach to emotional responses evoked through picture presentation
.
International Journal of Psychophysiology
,
72
,
212
216
.
JASP Team
. (
2018
).
JASP (Version 0.9) [Computer software]
. https://jasp-stats.org/.
Jensen
,
O.
,
Bonnefond
,
M.
, &
VanRullen
,
R.
(
2012
).
An oscillatory mechanism for prioritizing salient unattended stimuli
.
Trends in Cognitive Sciences
,
16
,
200
206
.
Jensen
,
O.
, &
Mazaheri
,
A.
(
2010
).
Shaping functional architecture by oscillatory alpha activity: Gating by inhibition
.
Frontiers in Human Neuroscience
,
4
,
186
.
Jessen
,
S.
, &
Grossmann
,
T.
(
2017
).
Exploring the role of spatial frequency information during neural emotion processing in human infants
.
Frontiers in Human Neuroscience
,
11
,
486
.
Junghöfer
,
M.
,
Elbert
,
T.
,
Tucker
,
D. M.
, &
Rockstroh
,
B.
(
2000
).
Statistical control of artifacts in dense array EEG/MEG studies
.
Psychophysiology
,
37
,
523
532
.
Kauffmann
,
L.
,
Chauvin
,
A.
,
Pichat
,
C.
, &
Peyrin
,
C.
(
2015
).
Effective connectivity in the neural network underlying coarse-to-fine categorization of visual scenes. A dynamic causal modeling study
.
Brain and Cognition
,
99
,
46
56
.
Kawasaki
,
H.
,
Adolphs
,
R.
,
Kaufman
,
O.
,
Damasio
,
H.
,
Damasio
,
A. R.
,
Granner
,
M.
, et al
(
2001
).
Single-neuron responses to emotional visual stimuli recorded in human ventral prefrontal cortex
.
Nature Neuroscience
,
4
,
15
16
.
Keil
,
A.
,
Moratti
,
S.
,
Sabatinelli
,
D.
,
Bradley
,
M. M.
, &
Lang
,
P. J.
(
2005
).
Additive effects of emotional content and spatial selective attention on electrocortical facilitation
.
Cerebral Cortex
,
15
,
1187
1197
.
Keitel
,
C.
,
Keitel
,
A.
,
Benwell
,
C. S.
,
Daube
,
C.
,
Thut
,
G.
, &
Gross
,
J.
(
2019
).
Stimulus-driven brain rhythms within the alpha band: The attentional-modulation conundrum
.
BioRxiv
,
336941
.
Kim
,
Y. J.
,
Grabowecky
,
M.
,
Paller
,
K. A.
,
Muthu
,
K.
, &
Suzuki
,
S.
(
2007
).
Attention induces synchronization-based response gain in steady-state visual evoked potentials
.
Nature Neuroscience
,
10
,
117
125
.
Klimesch
,
W.
(
2012
).
Alpha-band oscillations, attention, and controlled access to stored information
.
Trends in Cognitive Sciences
,
16
,
606
617
.
Kriegeskorte
,
N.
,
Simmons
,
W. K.
,
Bellgowan
,
P. S.
, &
Baker
,
C. I.
(
2009
).
Circular analysis in systems neuroscience: The dangers of double dipping
.
Nature Neuroscience
,
12
,
535
540
.
Kuffler
,
S. W.
(
1953
).
Discharge patterns and functional organization of mammalian retina
.
Journal of Neurophysiology
,
16
,
37
68
.
Lamme
,
V. A.
, &
Roelfsema
,
P. R.
(
2000
).
The distinct modes of vision offered by feedforward and recurrent processing
.
Trends in Neurosciences
,
23
,
571
579
.
LeDoux
,
J. E.
(
2000
).
Emotion circuits in the brain
.
Annual Review of Neuroscience
,
23
,
155
184
.
Li
,
Z.
,
Yan
,
A.
,
Guo
,
K.
, &
Li
,
W.
(
2019
).
Fear-related signals in the primary visual cortex
.
Current Biology
,
29
,
4078
4083
.
Liu
,
Y.
,
Keil
,
A.
, &
Ding
,
M.
(
2012
).
Effects of emotional conditioning on early visual processing: Temporal dynamics revealed by ERP single-trial analysis
.
Human Brain Mapping
,
33
,
909
919
.
Maris
,
E.
, &
Oostenveld
,
R.
(
2007
).
Nonparametric statistical testing of EEG- and MEG-data
.
Journal of Neuroscience Methods
,
164
,
177
190
.
Markovic
,
J.
,
Anderson
,
A. K.
, &
Todd
,
R. M.
(
2014
).
Tuning to the significant: Neural and genetic processes underlying affective enhancement of visual perception and memory
.
Behavioural Brain Research
,
259
,
229
241
.
Mather
,
M.
, &
Sutherland
,
M. R.
(
2011
).
Arousal-biased competition in perception and memory
.
Perspectives on Psychological Science
,
6
,
114
133
.
Maunsell
,
J. H.
, &
Treue
,
S.
(
2006
).
Feature-based attention in visual cortex
.
Trends in Neurosciences
,
29
,
317
322
.
McTeague
,
L. M.
,
Gruss
,
L. F.
, &
Keil
,
A.
(
2015
).
Aversive learning shapes neuronal orientation tuning in human visual cortex
.
Nature Communications
,
6
,
7823
.
Mermillod
,
M.
,
Guyader
,
N.
, &
Chauvin
,
A.
(
2005
).
The coarse-to-fine hypothesis revisited: Evidence from neuro-computational modeling
.
Brain and Cognition
,
57
,
151
157
.
Miller
,
K. D.
(
2003
).
Understanding layer 4 of the cortical circuit: A model based on cat V1
.
Cerebral cortex
,
13
,
73
82
.
Miskovic
,
V.
, &
Keil
,
A.
(
2012
).
Acquired fears reflected in cortical sensory processing: A review of electrophysiological studies of human classical conditioning
.
Psychophysiology
,
49
,
1230
1241
.
Moratti
,
S.
, &
Keil
,
A.
(
2005
).
Cortical activation during Pavlovian fear conditioning depends on heart rate response patterns: An MEG study
.
Cognitive Brain Research
,
25
,
459
471
.
Moratti
,
S.
,
Keil
,
A.
, &
Miller
,
G. A.
(
2006
).
Fear but not awareness predicts enhanced sensory processing in fear conditioning
.
Psychophysiology
,
43
,
216
226
.
Morris
,
J. S.
,
Friston
,
K. J.
,
Büchel
,
C.
,
Frith
,
C. D.
,
Young
,
A. W.
,
Calder
,
A. J.
, et al
(
1998
).
A neuromodulatory role for the human amygdala in processing emotional facial expressions
.
Brain
,
121
,
47
57
.
Movshon
,
J. A.
,
Thompson
,
I. D.
, &
Tolhurst
,
D. J.
(
1978
).
Spatial and temporal contrast sensitivity of neurones in areas 17 and 18 of the cat's visual cortex
.
Journal of Physiology
,
283
,
101
120
.
Müller
,
M. M.
,
Andersen
,
S. K.
, &
Keil
,
A.
(
2007
).
Time course of competition for visual processing resources between emotional pictures and foreground task
.
Cerebral Cortex
,
18
,
1892
1899
.
Müller
,
M. M.
,
Picton
,
T. W.
,
Valdes-Sosa
,
P.
,
Riera
,
J.
,
Teder-Sälejärvi
,
W. A.
, &
Hillyard
,
S. A.
(
1998
).
Effects of spatial selective attention on the steady-state visual evoked potential in the 20–28 Hz range
.
Cognitive Brain Research
,
6
,
249
261
.
Müller
,
M. M.
,
Teder
,
W.
, &
Hillyard
,
S. A.
(
1997
).
Magnetoencephalographic recording of steadystate visual evoked cortical activity
.
Brain Topography
,
9
,
163
168
.
Müller
,
M. M.
,
Teder-Sälejärvi
,
W.
, &
Hillyard
,
S. A.
(
1998
).
The time course of cortical facilitation during cued shifts of spatial attention
.
Nature Neuroscience
,
1
,
631
634
.
Palmer
,
S. E.
(
1999
).
Vision science: Photons to phenomenology
.
Cambridge, MA
:
MIT Press
.
Panitz
,
C.
,
Keil
,
A.
, &
Mueller
,
E. M.
(
2019
).
Extinction-resistant attention to long-term conditioned threat is indexed by selective visuocortical alpha suppression in humans
.
bioRxiv
,
533141
.
Pelli
,
D. G.
, &
Farell
,
B.
(
1999
).
Why use noise?
Journal of the Optical Society of America A
,
16
,
647
653
.
Pessoa
,
L.
, &
Adolphs
,
R.
(
2010
).
Emotion processing and the amygdala: From a “low road” to “many roads” of evaluating biological significance
.
Nature Reviews Neuroscience
,
11
,
773
783
.
Petro
,
N. M.
,
Gruss
,
L. F.
,
Yin
,
S.
,
Huang
,
H.
,
Miskovic
,
V.
,
Ding
,
M.
, et al
(
2017
).
Multimodal imaging evidence for a frontoparietal modulation of visual cortex during the selective processing of conditioned threat
.
Journal of Cognitive Neuroscience
,
29
,
953
967
.
Peyk
,
P.
,
De Cesarei
,
A.
, &
Junghöfer
,
M.
(
2011
).
Electro magneto encephalograhy software: Overview and integration with other EEG/MEG toolboxes
.
Computational Intelligence and Neuroscience
,
2011
,
861705
.
Pfurtscheller
,
G.
, &
Da Silva
,
F. L.
(
1999
).
Event-related EEG/MEG synchronization and desynchronization: Basic principles
.
Clinical Neurophysiology
,
110
,
1842
1857
.
Posner
,
M. I.
(
2008
).
Measuring alertness
.
Annals of the New York Academy of Sciences
,
1129
,
193
199
.
Pourtois
,
G.
,
Schettino
,
A.
, &
Vuilleumier
,
P.
(
2013
).
Brain mechanisms for emotional influences on perception and attention: What is magic and what is not
.
Biological Psychology
,
92
,
492
512
.
Regan
,
D.
(
1989
).
Human brain electrophysiology: Evoked potentials and evoked magnetic fields in science and medicine
.
New York
:
Elsevier
.
Rhodes
,
L. J.
,
Ruiz
,
A.
,
Ríos
,
M.
,
Nguyen
,
T.
, &
Miskovic
,
V.
(
2018
).
Differential aversive learning enhances orientation discrimination
.
Cognition and Emotion
,
32
,
885
891
.
Rihs
,
T. A.
,
Michel
,
C. M.
, &
Thut
,
G.
(
2007
).
Mechanisms of selective inhibition in visual spatial attention are indexed by α-band EEG synchronization
.
European Journal of Neuroscience
,
25
,
603
610
.
Ringach
,
D. L.
,
Bredfeldt
,
C. E.
,
Shapley
,
R. M.
, &
Hawken
,
M. J.
(
2002
).
Suppression of neural responses to nonoptimal stimuli correlates with tuning selectivity in macaque V1
.
Journal of Neurophysiology
,
87
,
1018
1027
.
Ringach
,
D. L.
,
Hawken
,
M. J.
, &
Shapley
,
R.
(
1997
).
Dynamics of orientation tuning in macaque primary visual cortex
.
Nature
,
387
,
281
284
.
Rosenthal
,
R.
, &
Rosnow
,
R. L.
(
1985
).
Contrast analysis: Focused comparisons in the analysis of variance
.
Cambridge, United Kingdom
:
Cambridge University Press
.
Rosnow
,
R. L.
,
Rosenthal
,
R.
, &
Rubin
,
D. B.
(
2000
).
Contrasts and correlations in effect-size estimation
.
Psychological Science
,
11
,
446
453
.
Sadaghiani
,
S.
, &
Kleinschmidt
,
A.
(
2016
).
Brain networks and α-oscillations: Structural and functional foundations of cognitive control
.
Trends in Cognitive Sciences
,
20
,
805
817
.
Saenz
,
M.
,
Buracas
,
G. T.
, &
Boynton
,
G. M.
(
2002
).
Global effects of feature-based attention in human visual cortex
.
Nature Neuroscience
,
5
,
631
632
.
Sagi
,
D.
, &
Tanne
,
D.
(
1994
).
Perceptual learning: Learning to see
.
Current Opinion in Neurobiology
,
4
,
195
199
.
Schyns
,
P. G.
, &
Oliva
,
A.
(
1994
).
From blobs to boundary edges: Evidence for time- and spatial-scale-dependent scene recognition
.
Psychological science
,
5
,
195
200
.
Serences
,
J. T.
,
Shomstein
,
S.
,
Leber
,
A. B.
,
Golay
,
X.
,
Egeth
,
H. E.
, &
Yantis
,
S.
(
2005
).
Coordination of voluntary and stimulus-driven attentional control in human cortex
.
Psychological Science
,
16
,
114
122
.
Skottun
,
B. C.
(
2015
).
On the use of spatial frequency to isolate contributions from the magnocellular and parvocellular systems and the dorsal and ventral cortical streams
.
Neuroscience & Biobehavioral Reviews
,
56
,
266
275
.
Snyder
,
A. C.
, &
Foxe
,
J. J.
(
2010
).
Anticipatory attentional suppression of visual features indexed by oscillatory alpha-band power increases: A high-density electrical mapping study
.
Journal of Neuroscience
,
30
,
4024
4032
.
So
,
Y. T.
, &
Shapley
,
R.
(
1981
).
Spatial tuning of cells in and around lateral geniculate nucleus of the cat: X and Y relay cells and perigeniculate interneurons
.
Journal of Neurophysiology
,
45
,
107
120
.
Song
,
I.
, &
Keil
,
A.
(
2014
).
Differential classical conditioning selectively heightens response gain of neural population activity in human visual cortex
.
Psychophysiology
,
51
,
1185
1194
.
Tallon-Baudry
,
C.
, &
Bertrand
,
O.
(
1999
).
Oscillatory gamma activity in humans and its role in object representation
.
Trends in Cognitive Sciences
,
3
,
151
162
.
Theeuwes
,
J.
(
1994
).
Endogenous and exogenous control of visual selection
.
Perception
,
23
,
429
440
.
Thigpen
,
N. N.
,
Bartsch
,
F.
, &
Keil
,
A.
(
2017
).
The malleability of emotional perception: Short-term plasticity in retinotopic neurons accompanies the formation of perceptual biases to threat
.
Journal of Experimental Psychology: General
,
146
,
464
471
.
Vagnoni
,
E.
,
Lourenco
,
S. F.
, &
Longo
,
M. R.
(
2015
).
Threat modulates neural responses to looming visual stimuli
.
European Journal of Neuroscience
,
42
,
2190
2202
.
Vuilleumier
,
P.
,
Armony
,
J. L.
,
Driver
,
J.
, &
Dolan
,
R. J.
(
2003
).
Distinct spatial frequency sensitivities for processing faces and emotional expressions
.
Nature Neuroscience
,
6
,
624
631
.
Vuilleumier
,
P.
, &
Driver
,
J.
(
2007
).
Modulation of visual processing by attention and emotion: Windows on causal interactions between human brain regions
.
Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences
,
362
,
837
855
.
Ward
,
J.
,
Rothen
,
N.
,
Chang
,
A.
, &
Kanai
,
R.
(
2017
).
The structure of inter-individual differences in visual ability: Evidence from the general population and synaesthesia
.
Vision Research
,
141
,
293
302
.
Watt
,
R. J.
(
1987
).
Scanning from coarse to fine spatial scales in the human visual system after the onset of a stimulus
.
Journal of the Optical Society of America A
,
4
,
2006
2021
.
Wieser
,
M. J.
, &
Keil
,
A.
(
2011
).
Temporal trade-off effects in sustained attention: Dynamics in visual cortex predict the target detection performance during distraction
.
Journal of Neuroscience
,
31
,
7784
7790
.
Xing
,
D.
,
Ringach
,
D. L.
,
Shapley
,
R.
, &
Hawken
,
M. J.
(
2004
).
Correlation of local and global orientation and spatial frequency tuning in macaque V1
.
Journal of Physiology
,
557
,
923
933
.
Zhang
,
W.
, &
Luck
,
S. J.
(
2009
).
Feature-based attention modulates feedforward visual processing
.
Nature Neuroscience
,
12
,
24
25
.
Zhu
,
W.
,
Xing
,
D.
,
Shelley
,
M.
, &
Shapley
,
R.
(
2010
).
Correlation between spatial frequency and orientation selectivity in V1 cortex: Implications of a network model
.
Vision Research
,
50
,
2261
2273
.