Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Q1 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Q3 55 56 57 58 59 60 61 Q2 62 63 64 65 66

Contents lists available at ScienceDirect

Neuropsychologia journal homepage: www.elsevier.com/locate/neuropsychologia

Low-contrast response deficits and increased neural noise in children with autism spectrum disorder Paige Mariel Weinger a,n, Vance Zemon c, Latha Soorya b, James Gordon d a Icahn School of Medicine at Mount Sinai Seaver Autism Center, Psychiatry One Gustave Levy Place, Box 1230 Atran Building E Level, Room 22, New York, NY 10029, United States b Rush University Medical Center, Chicago, Illinois, United States c Yeshiva University, New York, United States d Hunter College, New York, United States

art ic l e i nf o

a b s t r a c t

Article history: Received 3 December 2013 Received in revised form 13 June 2014 Accepted 28 July 2014

A battery of short-duration neurophysiological tests were designed and implemented using visual evoked potentials (VEPs) to examine specific neural mechanisms in children with and without autism spectrum disorder (ASD). Contrast-sweep conditions (bright or dark isolated-checks) were used to elicit steady-state VEPs to examine the integrity of ON/OFF pathways. Children with ASD displayed deficits in low-contrast responses at the stimulus frequency of 12.5 Hz, notably under conditions that emphasized activity in the magnocellular pathway. Signal-to-noise ratios were weaker in the ASD group, particularly for the OFF pathway. There were no group differences in the amplitude of responses. In addition, the ASD group displayed significantly higher levels of neural noise than controls. For the response at the stimulus frequency, the ASD group produced a relatively constant level of noise across the contrast range tested, with higher levels than controls at low contrasts and approximately equal levels of noise at moderate to high contrasts. & 2014 Published by Elsevier Ltd.

Keywords: Autism spectrum disorder Visual evoked potential Early-stage visual processing ON and OFF pathways Neural noise

1. Introduction Autism spectrum disorder (ASD) is a lifelong neurodevelopmental disorder that is often diagnosed during early childhood and is characterized by impairments in social communication, reciprocal social interaction, and repetitive or restricted behaviors and interests (American Psychiatric Association, 2013). Sensory symptoms have also been described as a hallmark of the disorder through both empirical and anecdotal accounts (Grandin & Scariano, 1986; Kern et al. 2006; O'Neill & Jones, 1997). Given the high prevalence of hyper- and hypo-sensitivities (Leekam, Nieto, Libby, Wing & Gould, 2007), sensory symptoms have been incorporated into the DSM-5 criteria for ASD (American Psychiatric Association, 2013). Electrophysiological studies of the visual system in individuals with ASD have revealed abnormalities in both low-level (Vandenbroucke, Scholte, van Engeland, Lamme, & Kemner, 2008) as well as higher-level visual pathways (McPartland, Coffman, & Pelphrey, 2011). Although there is evidence to suggest that an underlying basic sensory problem may have a direct impact on the signs and symptoms of ASD, many

n

Corresponding author. Tel.: þ 1 440112122417250. E-mail addresses: [email protected], [email protected] (P.M. Weinger).

studies focus on later visual processes that are known to be associated with high-level perception (e.g., social-cognitive deficits), and fewer studies examine low-level sensory processing. In order to better understand the altered perceptual functioning found in many individuals with ASD, the early stages of sensory processing, including neural responses to basic non-social stimuli must be examined further (McPartland et al., 2011; Simmons et al., 2009). Recording visual evoked potentials (VEPs) are a noninvasive means to examine the integrity of specific brain mechanisms that may underlie the disorder. Through an analysis of real-time brain activity, reflected in the electroencephalogram (EEG), with a temporal resolution on the order of milliseconds, information about summed excitatory and inhibitory postsynaptic potentials can be gathered quickly (Zemon, Kaplan & Ratliff, 1980, 1986). In the current study, contrast response functions were obtained to examine several aspects of early-stage visual processing in children with and without ASD. Specifically, low-contrast bright and dark isolated-check stimuli were presented to selectively tap ON and OFF cells in the magnocellular pathway (Zemon & Gordon, 2006). The magnocellular pathway is made up of the two ventral layers of neurons in the dorsal lateral geniculate nucleus of the thalamus (dLGN) and plays a critical role in the perception of form, movement, depth and brightness (Kaplan, 2004). The cortical recipients of magnocellular input signals are responsive to low

http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031 0028-3932/& 2014 Published by Elsevier Ltd.

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

2

1 2 Q4 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

contrast stimuli and exhibit low spatial resolution and high temporal resolution (Lee, Pokorny, Smith, Martin & Valberg, 1990; Tootell, Hamilton, & Switkes, 1988; Tootell, Silverman, Hamilton, Switkes & De Valois, 1988). ON and OFF pathways constitute distinct parallel subsystems that transmit positive- and negative-contrast information to the visual cortex, and the resulting cortical activation forms the basis of brightness and darkness perception (Schiller, 1982; Zemon, Gordon & Welch, 1988). When light hits the central field, ON cells are excited, and when light hits the surrounding field, OFF cells are excited (Schiller, Sandell & Maunsell, 1986). Therefore, light objects on dark backgrounds are detected via ON cells and dark objects on light backgrounds are detected via OFF cells. Isolated-check VEPs use these principles to assess select neural pathways in the visual system by examining central vision through the display of a single contrast-polarity stimulus pattern. Furthermore, given that ON and OFF cells are considered distinct components of the visual system, sensory information is likely processed in a differential manner (De Valois, 1977; Magnussen & Glad, 1975; Zemon et al. 1988). In order to emphasize magnocellular pathway contributions, stimuli are varied in the low-contrast region at moderate to high temporal frequencies (Zemon & Gordon, 2006). Previous studies have used bright and dark isolated-check stimuli to examine developmental effects on responses to temporally-modulated spatial patterns. Findings indicate that maturational changes occur in both magnitude and phase of response frequency components throughout the lifespan; the developmental time course within ON and OFF pathways appears to be similar (Zemon et al. 1995). Findings in typically developing children indicate that response amplitude decreases with age above the age of six (Dustman & Beck, 1966, 1969; Shaw & Cant, 1981); this finding holds under both bright and dark-check conditions (Zemon et al., 1995). In addition, the dark-check condition produces larger responses than does the bright-check condition in young infants (Hartmann, Hitchcox, & Zemon, 1992), similar to that found in adults. According to Zemon et al. (1995), these findings may be attributed to changes in cortical circuitry. Specifically, the decrease in amplitude is thought to reflect the loss of excitatory synaptic activity that occurs with increasing age (Fosse, Heggelund, & Fonnum, 1989; Rakic, Bourgeois, Eckenhoff, Zecevic, & Goldman-Rakic 1986; Zemon et al., 1995). While bright and dark isolated-check conditions have not yet been applied to the study of early-stage visual processing in ASD, several studies have attempted to examine the integrity of the magnocellular pathway in individuals with ASD, high-risk infants, and in the broader autism phenotype. A recent study by Greenaway, Davis, and Plaisted-Grant (2013) found significantly higher contrast discrimination thresholds on the steady-pedestal condition of the Pokorny and Smith task (Pokorny & Smith, 1997) in a sample of high-functioning children with ASD. The steadypedestal condition is thought to reflect magnocellular activity, while the pulsed-pedestal condition is thought to reflect parvocellular activity. The cortical recipients of parvocellular input signals are responsive to high contrast stimuli and exhibit high spatial resolution (Lee et al., 1990; Tootell, Hamilton et al., 1988; Tootell, Silverman et al., 1988). No group differences were found in response to the pulsed-pedestal condition, thus indicating selective magnocellular impairment in children with ASD. McCleary, Allman, Carver and Dobkins (2007) applied luminance contrast sensitivity methods to a sample of infant siblings of children with ASD. This high-risk sample displayed thresholds that were almost double (i.e., lower sensitivities) those of controls in a condition emphasizing the magnocellular pathway, while there were no group differences in a condition emphasizing the parvocellular pathway. Sutherland and Crewther (2010) used electrophysiological and psychophysical measures to examine the visual system in

typically developing adults with high vs. low scores on the Autism Spectrum Quotient (AQ; Baron-Cohen, Wheelwright, Skinner, Martin & Clubley, 2001). Higher scores reflect a greater number of ASD symptoms, thus capturing the broader autism phenotype. Results indicated that high scorers displayed weaker initial cortical responses at low contrasts and poorer performance on Navon figures that assess the ability to distinguish global aspects of figures. Overall, the authors argued that results may reflect deficits in global visual perception in individuals with ASD. In comparison, Bertone, Mottron, Jelenic, and Faubert (2005) examined orientation discrimination using a flicker contrast sensitivity task. Results indicated better orientation discrimination for first-order, luminance-defined gratings and decreased orientation discrimination for second-order, texture-defined gratings in ASD participants. The authors attributed the dichotomy of enhanced and diminished processing of visual-spatial information to abnormal connectivity and enhanced lateral inhibition, not magnocellular or parvocellular function. Koh, Milne, and Dobkins (2010) also reported no difference in response to a luminance-defined contrast sensitivity task. Fujita, Yamasaki, Kamio, Hirose, and Tobimatsu (2011) identified subcortical deficits through an examination of the magnocellular and parvocellular pathways in adolescents and adults with ASD using 128-channel high-density EEG recording. An analysis of peaks and troughs in the VEP waveforms indicated longer N1 (100 ms) latencies in the ASD group with chromatic, but not achromatic stimulation. Results were indicative of impaired parvocellular and intact magnocellular functioning in ASD. While conflicting results have been reported, stimulus parameters significantly impact which pathways are being targeted. Neumann et al. (2010) evaluated both early- and late-stage Q5 visual processing in 10 high functioning adolescents and adults with ASD using an adapted version of the Embedded Figures Task (Mottron, Burack, Iarocci, Belleville, and Enns, 2003). The task required participants to determine whether the letter “S” or “H” was embedded or isolated in a pattern stimulus. Results indicated no differences between ASD and control participants in behavioral performance (accuracy and reaction time). At early stages of processing (100–150 ms), differences were found in magnetoencephalographic (MEG) responses for the embedded condition, but not for the isolated condition. At later time intervals (350– 400 ms), amplitude differences were found between groups for all conditions. In addition, an analysis of source localization showed peaks in brain activity in the primary visual cortex of ASD participants for all conditions and time windows, while brain activity in the control participants peaked in other cortical areas. The authors argued that the enhanced activation seen in V1 of ASD participants provides support for the Enhanced Perceptual Functioning theory (Mottron, Dawson, Soulieres, Hubert, and Burack 2006). The differences seen in early time domains may be indicative of reduced processing when the context is irrelevant. Furthermore, differences in the location of responses are said to reflect bottom-up processing in ASD participants (Neumann et al., 2010). Baruth, Casanova, Sears, and Sokhadez (2010) examined early- Q6 stage visual processing in high functioning children and adolescents with ASD (ages 9–20) using event-related potentials (ERPs) to an oddball visual illusory task. Results indicate that P50 amplitudes in parieto-occipital regions of interest were significantly more positive in the ASD group in response to non-target stimuli, while P50 latencies were significantly reduced. In general, the ASD group displayed abnormally large responses to task irrelevant stimuli, particularly in parieto-occipital and frontal regions of interest. Behavioral results indicated no group differences in reaction time; however, ASD participants displayed difficulty with stimulus discrimination and had a greater number of motor response errors. The authors argue that weak inhibitory

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

input in the cortex may result in impaired visual input regulation. Vandenbroucke et al. (2008) also used ERPs to examine low-level visual processing in an older sample of teenagers and young adults (ages 16–28). Results from an object boundary detection task demonstrated that group differences were present at approximately 120 ms, but not at 260 ms. These findings were described as a reflection of abnormal horizontal connectivity in early stages of visual processing, with compensatory mechanisms accounting for intact processing at later stages. The authors argue that deficits in visual perception may be a result of impaired lateral inhibition (Vandenbroucke et al., 2008). Taken together, it is evident that results examining low level visual processing remain inconclusive. The variability in methodology used to examine specific functions within the visual cortex, in combination with small sample sizes, makes it difficult to compare results across studies. However, there is evidence to suggest that additional research is needed to examine low-level perception in ASD. The current study also examines neural noise by assessing variability in VEP responses. Studies across a variety of disciplines have provided evidence of increased neural noise in ASD (see review by Simmons et al. (2009)). The increase in noise has been described in the context of both hyperexcitation in key sensory areas and as a result of decreased inhibition (Merzenich, Saunders & Tallal, 1999; Merzenich, Tallal, Peterson, Miller & Jenkins, 1998; Rubenstein & Merzenich, 2003). At the cellular level, individuals with ASD have been found to have fewer Purkinje cells when compared to controls (Bauman & Kemper, 1985, 1994), and thus abnormal excitability and increased cortical noise may be a result of the inhibitory influence of such cells (Baron-Cohen & Belmonte, 2005; Courchesne, 1997). In addition to findings from cellular and anatomical studies, neural abnormalities have also been reported across sensory systems using functional measures such as EEG and fMRI. In a study examining intra-participant variability in the EEG responses of children and adolescents using Gabor patches, greater variability in P1 latencies and amplitudes in the ASD group was attributed to an increase in neural noise and impairments in the synchronization of cell assemblies based on particular stimuli (Milne, 2011). Dinstein et al. (2010) and Dinstein, Heeger, Minshew, Malach, and Behrmann (2012) also reported increased response variability in two fMRI studies of high functioning adults with ASD. In the initial study (Dinstein et al., 2010), within-subject variability was identified in the ASD group during a task examining the mirror neuron system based on observed and executed hand movements. Results indicated no difference in response amplitudes between groups, and thus intact mirror neuron system function; however, variability in neural responses within subjects was identified in the ASD group relative to controls. In a more recent study (Dinstein et al., 2012), fMRI was used to examine response variability across visual, auditory, and somatosensory cortices. Results indicated increased response variability in the ASD group for evoked cortical responses to low-level stimuli in both the visual and somatosensory cortex. Weaker signal-to-noise ratios were also identified in the ASD group across all three sensory cortices examined. There were no significant differences in mean amplitudes or response variability during resting-state activity. These results support theories suggesting that individuals with ASD display global neural processing abnormalities (Minshew, Goldstein, & Siegel, 1997; Belmonte et al., 2004), which may underlie differences in cognitive factors associated with the disorder (Belmonte et al., 2004). Overall, results suggest that continued examination of the relationship between abnormal sensory processing and levels of neural noise is warranted. The current study takes advantage of the rapid, reliable assessment provided by VEP recording as a means of examining distinct neural mechanisms within the visual pathways of children

3

with ASD. Primary aims include an examination of (1) cortical activation via ON- and OFF-cell pathways within the magnocellular system and (2) the presence of neural noise within the visual cortex. We hypothesize that children with ASD will show deficits in the magnocellular pathway and increased within-subject response variability within the primary visual cortex in response to steady-state isolated-check VEPs.

2. Materials and methods 2.1. Participants Twenty-four children with ASD (4 females, 20 males, M ¼ 7.88 yr, SD ¼2.54) and 18 typically developing children (7 females, 11 males, M ¼ 7.50 yr, SD ¼2.36) participated in this study. ASD participants were diagnosed according to a consensus diagnosis determined by DSM-IV-TR (APA, 2000) or DSM-5 (APA, Q7 2013) criteria, a clinical intake with a licensed psychologist or child and adolescent psychiatrist, and standardized assessments including the Autism Diagnostic Observation Schedule (ADOS; Lord, Rutter, DiLavore, & Risi, 1999) and the Autism Diagnostic Interview-Revised (ADI-R; Rutter, LeCouteur & Lord, 2003). All diagnostic assessments were administered and scored by research reliable raters. Intellectual functioning was measured in the ASD group using the Wechsler Intelligence Scale for Children, Fourth Edition (Wechsler, 2003), the Stanford Binet, Fifth Edition (Roid, 2003), the Differential Ability Scales, Second Edition (Elliot, 2007), or the Mullen Scales of Early Learning (Mullen, 1995). All participants were screened with the Social Responsiveness Scale (Constantino & Gruber, 2005). Visual acuity was measured both monocularly and binocularly at 20 ft and 114 cm using a Snellen chart. All participants had normal (20/20) or corrected to normal acuity from the viewing distance of 114 cm and from 20 ft. Informed written consent was obtained on all participants and the internal review board of the participating universities approved the experiments. Data from three ASD participants were removed because of noise artifacts (n ¼2) or because they did not meet ASD criteria on standardized diagnostic assessments (n¼ 1). Data from four control participants were removed because total scores on the SRS were above the ASD cut-off (n¼ 2), there were noise artifacts (n ¼1), or because of poor visual acuity (n ¼1). The final sample was comprised of 21 children with ASD (M ¼ 8.00 yr, SD ¼2.53) and 14 control participants (M ¼ 7.57 yr, SD ¼ 2.24). An independent-samples t-test indicated no difference in age (t(31) ¼  .513, p ¼.603). Results from a chi-square test indicated an association between group and sex (p¼ .022). Total scores on the Social Responsiveness Scale differed significantly by group (p o .001). Participant characteristics for the ASD group are shown in Table 1. 2.2. VEP recording A Neucodia system (VeriSci Corp.) was used for stimulus presentation, data collection and analysis. Gold-cup electrodes were placed on the midline of the scalp based on the 10–20 system (Jasper, 1958), which includes an active electrode at Oz (occipital), a reference electrode at Cz (vertex), and a ground electrode at Pz (in between Oz and Cz). These three electrodes comprised a single electrophysiological channel. All EEG samples were recorded synchronized to the display's frame rate (4 samples per frame). The Neucodia system provided automated noise and outlier detection. The noise detection feature determined whether the EEG recording was affected by 60 Hz noise, drift, or saturation. Artifacts such as eyeblinks were detected by automated artifact rejection features built in to the software, which rejected EEG epochs that contained large potential spikes or drifts in the recording. The multivariate outlier analysis feature detected extreme values in the set of responses relative to the other responses, based on a statistical criterion of .05 significance level. Similar to the noise detection feature, outliers Table 1 Participant characteristics for the ASD sample. IQ scores are listed as standard scores. ASD participant characteristics Full scale IQ (M, SD) Verbal IQ (M, SD) Nonverbal IQ (M, SD) ADOS social affect (M, SD) ADOS repetitive restricted behaviors (M, SD) ADOS total (M, SD) ADI social (M, SD) ADI communication (M, SD) ADI Interests and Behaviors (M, SD)

93.05 93.68 98.32 8.75 3.15 11.95 19 15.17 6.56

(29.09) (30.95) (29.84) (2.90) (1.60) (3.50) (7.42) (5.79) (3.22)

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Fig. 1. Dark and bright isolated-check patterns. were rejected and the examiner was prompted to repeat runs until 10 valid runs were collected. Fixation was monitored by means of an infrared camera/source and a display monitor, which was dedicated exclusively for fixation tracking and monitored throughout all runs. Any observance of looking away from the center of the display resulted in the experimenter rejecting that EEG epoch. At the end of each run, the system asked the examiner about fixation (e.g., “Was the participant looking at the screen?”) to ensure good quality data. The EEG was amplified (gain ¼20,000, bandpass filter: .5–100 Hz), digitized (600 samples/s), and stored on the computer. 2.3. Stimuli Stimulus conditions were presented with a field size that subtended 101  101 of visual angle (viewing distance¼114 cm). Background luminance was  50 cd/m2 and the frame rate was 150 Hz. Stimuli included contrast sweeps of 16  16 bright or dark isolated-checks with appearance/disappearance sinusoidal modulation (Fig. 1). Depth of modulation increased in octave steps from 1–32% (peak contrast: 2–64%) at 12.5 Hz. Each swept-parameter condition consisted of 10 7-s runs with  1-s of adaptation included at the beginning of each run. 2.4. Analysis A discrete Fourier transform was performed to extract the harmonic frequency components in the response. Given that under these stimulus conditions the dominant response component occurs at the fundamental frequency (i.e., stimulus frequency), all analyses were focused on this component. Each of the 10 individual responses was characterized by the sine and cosine coefficients of the relevant frequency component, and mean amplitude and phase values were calculated. A 95% confidence circle around the mean vector response was computed, based on the T 2circ statistic (Victor & Mast, 1991), in order to specify an estimate of variability in amplitude (amplitude error, noise) and phase (phase error). Noise is defined by the radius of the 95% confidence circle. Multivariate statistics were used because of the nature of the vector responses collected, which include both sine and cosine coefficients (yielding amplitude and phase measures). In order to capture the variability of the response, both variables (amplitude and phase) are needed. In addition, this statistic was used to calculate signal-to-noise ratios (SNRs), with a value 41 indicating a response out of the noise at the .05 level of significance. Multivariate ANOVAs were used to assess overall differences between groups. For measures of signal-to-noise and amplitude, ANOVAs included responses based on all DOMs above 2%, which yielded mean responses above the noise level for the control group. Bootstrap analyses were used to ensure robust results given the limitations of the sample.

3. Results 3.1. Signal-to-noise ratio Multivariate ANOVA was used to examine group differences and Group  DOM interactions. The ASD group displayed significantly weaker SNRs than did the control group for the dark-check condition (F(4, 29) ¼ 3.38, p ¼.02, ηp²¼ .318 ) and approached significance for the bright-check condition (F(4, 30) ¼2.66, p ¼.052, ηp² ¼.262). For the dark-check condition, there was a

Fig. 2. Mean signal-to-noise ratios (SNR) were plotted as a function of depth of modulation (DOM) for the dark-check condition. SNR was calculated by dividing the mean amplitude of the fundamental response by the radius of the 95% confidence circle. SNRs 41.0 exceed background noise. Error bars are 7 1 SE.

significant difference between groups at 4% DOM (p ¼.015, ηp² ¼.173), 8% DOM (p ¼.001, ηp² ¼.280), 16% DOM (p ¼.027, .143), and 32% DOM (p ¼.049, ηp² ¼.115). For the bright-check condition, there was a significant difference between groups at 4% DOM (p ¼.002, ηp² ¼.248) and 8% DOM (p¼ .046, ηp² ¼.115). The difference was not significant at 16% DOM (p¼ .098, ηp²¼.081) or at 32% DOM (p ¼.055, ηp² ¼.107). Results from a bootstrap technique based on 1000 samples indicated a significant difference in the dark-check condition at 4% DOM (p ¼.025), 8% DOM (p ¼.001), and 16% (p ¼.036). The difference was not significant at 32% DOM (p¼ .063). For the bright check condition, there was a significant difference between groups at 4% DOM (p¼ .024). The difference was not significant at 8% DOM (p ¼.056), 16% DOM (p¼ .117), or 32% (p ¼.083). Results are displayed in Figs. 2 and 3. Repeated measures ANCOVAs were run to assess the effects of age and sex. In the dark-check condition, there was no significant main effect for age (F(1, 32) ¼1.07, p¼ .308, ηp² ¼.032) or sex

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

5

in the magnitude of cortical responses. Multivariate ANOVA was used to examine group differences and Group  DOM interactions. There were no significant differences in amplitude for the darkcheck condition (F(4, 29) ¼.68, p ¼.611, ηp²¼ .068) or the brightcheck condition (F(4, 30)¼ 1.24, p¼ .317, ηp² ¼.141) (Figs. 4 and 5). Furthermore, there were no significant differences between groups at any DOM, which is consistent with the large variability in amplitudes depicted in Figs. 6 and 7. Results from a bootstrap technique also indicated no significant differences between groups at any DOM. In the dark-check condition, there was no significant

Fig. 3. Mean signal-to-noise ratios (SNRs) were plotted as a function of depth of modulation (DOM) for the bright-check condition. SNR was calculated by dividing the mean amplitude of the fundamental response by the radius of the 95% confidence circle. SNRs 4 1.0 exceed background noise. Error bars are 71 SE.

Fig. 5. Mean amplitude plotted as a function of depth of modulation (DOM) for bright-check conditions. Error bars are 7 1 SE.

Fig. 4. Mean amplitude plotted as a function of depth of modulation (DOM) for dark-check conditions. Error bars are 7 1 SE.

(F(1, 32) ¼.826, p ¼.370, ηp² ¼.025). Similarly, there was no significant effect for age (F(1, 33)¼.149, p¼ .702, ηp² ¼.004) or sex (F(1, 33)¼ 2.71, p ¼.110, ηp²¼ .076) in the bright-check condition. 3.2. Amplitude Amplitudes of the fundamental harmonic component were examined to determine whether differences exist between groups

Fig. 6. Amplitude for the dark-check condition plotted as a function of depth of modulation (DOM) for all participants.

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Fig. 7. Amplitude for the bright-check condition plotted as a function of depth of modulation (DOM) for all participants.

Fig. 9. Mean noise plotted as a function of depth of modulation (DOM) for brightcheck condition. Noise is measured in microvolts (mV) Error bars are 7 1 SE.

displayed significantly greater levels of noise at 1% DOM (p ¼.001, ηp² ¼.292), 2% DOM (p¼ .016, ηp² ¼.168), and 4% DOM (p¼ .026, ηp² ¼.145). For the bright-check condition, the ASD group displayed significantly greater levels of noise at 1% DOM (p¼ .031, ηp²¼ .134) and 4% DOM (p ¼ .04, ηp²¼ .122). The difference was not significant at 2% DOM (p ¼.061, ηp² ¼.102). There were no differences between groups at 8–32% DOM for either condition. The level of noise in the ASD group remained relatively constant across the DOMs, whereas the noise level in the control group was low at lower DOMs and increased at higher DOMs (Figs. 8 and 9). Results from a bootstrap technique based on 1000 samples indicated that within the dark-check condition, the difference between groups was significant at 1% DOM (p ¼.001), 2% DOM (p ¼.017), and 4% DOM (p ¼.031). For the bright-check condition, there was a significant difference between groups at 1% DOM (p ¼.028) and 4% DOM (p¼ .050). The difference was not significant at 2% DOM (p¼ .058). In the dark-check condition, there was no significant main effect for age (F(1, 32)¼ .019, p ¼ .892, ηp²¼ .001) or sex (F(1, 32) ¼ 2.037, p ¼.163, ηp² ¼.060). In the bright-check condition, there was no significant main effect for age (F(1, 32)¼ 1.136, p¼ .294, ηp² ¼.034) or sex (F(1, 33)¼3.794, p ¼.060, ηp² ¼.103).

Fig. 8. Mean noise plotted as a function of depth of modulation (DOM) for darkcheck condition. Noise is measured in microvolts (mV). Error bars are 7 1 SE.

main effect for age (F(1, 33)¼.401, p ¼.531, ηp² ¼.012) or sex (F(1, 33)¼ .178, p ¼.676, ηp²¼ .005). Similarly, for the bright-check condition, there was no significant effect for age (F(1, 32) ¼1.136, p ¼.294, ηp² ¼.034) or sex (F(1, 32) ¼.065, p ¼.800, ηp² ¼.002). 3.3. Noise Multivariate ANOVA indicated that the ASD group displayed significantly greater within-subject response variability (neural noise) for the dark-check condition (F(6, 27) ¼ 4.22, p ¼.004, ηp²¼ .484 ), but not the bright-check condition (F(6, 28) ¼2.275, p ¼.065, ηp² ¼.328). For the dark-check condition, the ASD group

4. Discussion Contrast response functions based on VEP data were used to examine early-stage visual processing in a sample of children with ASD. VEPs were obtained under isolated bright- and dark-check stimulation conditions to independently tap ON- and OFF-cell activity. In order to emphasize contributions of the magnocellular pathway, the stimuli used in this study were modulated at a moderately high temporal frequency and varied from low to high contrast. In addition, an analysis of amplitude and phase data measured on the fundamental frequency component of the response was used to assess the integrity of the underlying neural pathways, such as the amplitude of responses and response variability (neural noise).

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 Q8 15 16 17 18 19 20 Q9 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Signal-to-noise ratios were first examined to determine the magnitude of VEP responses relative to noise level in the EEG. The ASD group demonstrated significantly weaker SNRs compared to controls for the dark-check condition at all levels of contrast above 2% DOM. Significant differences were also found at 4% and 8% DOM in the bright-check condition. There were no differences between groups in the amplitude of responses for either condition. Children in both groups displayed the expected pattern in contrast response functions, in which amplitude rises and then saturates, while phase advances with increases in DOM. Consistent with previous findings, VEPs were larger (SNR and amplitude) in the dark-check condition for both groups, which indicates greater contrast gain in the OFF pathway as compared to the ON pathway (Zemon et al., 2006; Zemon et al., 1988). These findings also support previous work in the field of autism in which both electrophysiological and psychophysical measures provide evidence for abnormalities in the magnocellular pathway in individuals with ASD (Greenaway et al., 2013), high-risk infant siblings (McCleary et al., 2007) and in the broader autism phenotype (Sutherland et al., 2010). Our results provide electrophysiological support for recent findings by Greenaway et al. (2013), who reported impaired magnocellular function in a sample of children with ASD in response to a steady pedestal, psychophysical task that assessed luminance contrast sensitivity. In addition, higher motion coherence thresholds were reported in several ASD studies (Kaiser & Shiffrar, 2009; Milne et al. 2002; Spencer et al. 2000), which may reflect both magnocellular deficits and increased neural noise as described below. Our findings support selective deficits in ON- and OFF-cell pathways in children with ASD. However, due to the complexity and variety of GABAergic interneurons and the role they play in the cortex, it is possible that certain aspects of GABAergic function may be reduced, while other aspects may be heightened. This dichotomy was previously described by Bertone et al. (2005)) in which results from psychophysical tasks indicated superior orientation discrimination for first-order, luminance-defined, gratings, and inferior orientation discrimination for second-order, texturedefined, gratings in high-functioning individuals with ASD. One possibility for the difference in results obtained from studies using psychophysical tasks (e.g., flicker contrast sensitivity) and our electrophysiological task is the difference in temporal frequency used to modulate the stimuli. For example, Bertone et al. (2005) used 6 Hz, while we used 12.5 Hz, which is more likely to emphasize magnocellular activity as compared to lower temporal frequencies. Furthermore, psychophysical and VEP measures are fundamentally different: VEPs reflect a population response—sum of a large number of excitatory and inhibitory postsynaptic potentials in the visual cortex (Zemon et al. 1986), whereas a psychophysical response can be produced by activity in a small group of neurons that have intact function. It will be important for future studies to incorporate both psychophysical and electrophysiological assessments to determine levels of consistency between behavioral and neural responses. Furthermore, if there is a lack of input from the magnocellular pathway, there may be a lack of interplay via GABAergic interneurons between cortical neurons that typically receive inputs from magnocellular and parvocellular neurons. From a behavioral perspective, deficits in the magnocellular pathway may result in difficulty forming global percepts, which could result in a preference for local vs. global processing (Happé & Frith, 2006). In addition to magnocellular deficits, results from measures of amplitude error indicate that the ASD group had significantly greater neural noise than the control group. This finding was present at the lowest levels of contrast (1–4% DOM) for the darkcheck condition when SNRs were below one. In the bright-check condition, the ASD group displayed greater levels of noise than

7

controls at 1% and 4% DOM. Overall, results suggest greater response variability when no effective input to the cortex was being received (SNR o 1). At the response frequency of interest (fundamental component), the ASD group displayed a relatively constant level of noise across the range of contrasts used, whereas the control group displayed low levels of noise at low contrasts, followed by a rise in noise as contrast increased; this finding was present across stimulus conditions. Thus, group differences were significant only under lower contrast conditions in which responses were either absent or weak. Our findings provide objective physiological support for theories of increased neural noise in ASD. Results are consistent with Dinstein et al. (2012) who reported decreased signal-to-noise ratios, no difference in amplitudes, and increased response variability based on evoked responses in the visual cortex using fMRI. Furthermore, our findings contribute to the growing body of literature spanning anatomical, cellular, and functional studies which suggest that neural noise is increased in ASD (Baron-Cohen & Belmonte, 2005; Dinstein et al., 2012; Rubenstein et al., 2003; Simmons et al., Q10 2009). Future work is necessary to determine how this finding translates to phenotypic and genotypic factors. Our study demonstrates the feasibility of short-duration, singlechannel VEP recording in children at various levels of functioning and it is the first study to use isolated-check, steady-state VEPs in this population. Our findings suggest that VEPs hold promise as a rapid and reliable method to gather objective information about neural processing using low-level, non-social, stimuli in the absence of a behavioral task. However, this study was limited by several factors. Specifically, despite developmental changes taking place throughout childhood, the sample was comprised of a wide age range, which was not matched by sex. While statistical analyses indicated that there was no significant effect of sex or age in response to any of the stimuli, future studies would benefit from larger, better-matched, samples that allow for age-based subgroups. Additional variables to consider include level of cognitive and adaptive functioning, patterns of neural abnormality, and the examination of inter- vs. intra-participant variability. Given the high prevalence of sensory symptoms in children with ASD, particularly within the visual domain (Kern et al., 2001; Q11 O'Neill & Jones, 1997), there is a growing need to develop methods to objectively examine sensory functioning. In addition, results from parent- and self-report measures indicate that visual hyperand hypo-sensitivities may persist throughout the lifespan (Crane Q12 et al., 2009; Leekam et al., 2007). Although the severity of these symptoms often declines with age, reports indicating the presence of visual symptoms in adulthood suggest that further investigation is warranted to determine whether improvement in some individuals may be a result of compensatory factors or whether changes in low-level factors may be a result of neural plasticity. Furthermore, visual symptoms range from non-social (e.g., visual inspection, finger/object flicking in the visual periphery, sensitivity to lights) to social behavior (e.g., unusual eye contact, emotion recognition deficits), which raise a critical question as to whether the higher-level social and cognitive dysfunction observed in ASD may be predicated on deficits in low-level visual function.

Acknowledgments We would like to thank the children and families who participated in this study and the individuals who assisted with recruitment and data collection. A special thank you to Valerie Nunez, Theresa Navalta, Adeola Harewood, Stacey Lurie and Jesslyn Jaminson. This work was partially supported by Autism Speaks Q13 (Grant #8685).

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

8

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 References 2 3 American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (text revision). (4th ed.). Washington, DC: American Psychiatric 4 Association. 5 American Psychiatric Association. (2013). Diagnostic and statistical manual of mental 6 disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Baron-Cohen, S., & Belmonte, M. K. (2005). Autism: a window onto the develop7 ment of the social and the analytic brain. Annual Review of Neuroscience, 28, 8 109–126. 9 Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The 10 autism-spectrum quotient (AQ): evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of 11 Autism and Developmental Disorders, 31(1), 5–17. 12 Bauman, M. L., & Kemper, T. L. (1985). Histoanatomic observations of the brain in 13 early infantile autism. Neurology, 35, 866–874. Bauman, M. L., & Kemper, T. L. (1994). Neuroanatomic observations of the brain in 14 autism. The neurobiology of autism (pp. 119–145)Baltimore: Johns Hopkins 15 University Press. 16 Belmonte, M. K., Cook, E. H., Jr., Anderson, G. M., Rubenstein, J. L., Greenough, W. T., 17 Beckel-Mitchener, A., et al. (2004). Autism as a disorder of neural information processing: directions for research and targets for therapy. Molecular Psychiatry, 18 9(7), 646–663. 19 Bertone, A., Mottron, L., Jelenic, P., & Faubert, J. (2005). Enhanced and diminished 20 visuo-spatial information processing in autism depends on stimulus complexity. Brain, 128, 2430–2441. 21 Constantino, J. N., & Gruber, C. P. (2005). Social responsiveness scale (SRS). Los 22 Angeles: Western Psychological Services. 23 Courchesne, E. (1997). Brainstem, cerebellar and limbic neuroanatomical abnorm24 alities in autism. Current Opinion in Neurobiology, 7(2), 269–278. DeValois, K. K. (1977). Spatial frequency adaptation can enhance contrast sensitiv25 ity. Vision Research, 17(9), 1057–1065. 26 Dinstein, I., Heeger, D. J., Minshew, N. J., Malach, R., & Behrmann, M. (2012). 27 Unreliable evoked responses in autism. Neuron, 75(6), 981–991. Dinstein, I., Thomas, C., Humphreys, K., Minshew, N., Behrmann, M., & Heeger, D. J. 28 (2010). Normal movement selectivity in autism. Neuron, 66, 461–469. 29 Dustman, R. E., & Beck, E. C. (1966). Visually evoked potentials: amplitude changes 30 with age. Science, 151(25), 1013–1015. 31 Dustman, R. E., & Beck, E. C. (1969). The effects of maturation and aging on the wave form of visually evoked potentials. Electroencephalography Clinical Neurophy32 siology, 26(1), 2–11. 33 Elliott, C. D. (2007). Differential ability scales: introductory and technical handbook 34 (2nd edition). San Antonio, TX: The Psychological Corporation. Fosse, V. M., Heggelund, P., & Fonnum, F. (1989). Postnatal development of 35 glutamatergic, GABAergic, and cholinergic neurotransmitter phenotypes in 36 the visual cortex, lateral geniculate nucleus, pulvinar, and superior colliculus 37 in cats. Journal of Neuroscience, 9(2), 426–435. 38 Fujita, T., Yamasaki, T., Kamio, Y., Hirose, S., & Tobimatsu, S. (2011). Parvocellular pathway impairment in autism spectrum disorder: evidence from visual 39 evoked potentials. Research in Autism Spectrum Disorders, 5(1), 277–285. 40Q14 Grandin, T., & Scariano, M. (1986). Emergence: labeled autistic novato. . 41 Greenaway, R., Davis, G, & Plaisted-Grant, K. (2013). Marked selective impairment in autism on an index of magnocellular function. Neuropsychologia, 51, 42 592–600. 43 Q15 Happé, F., & Frith, U. (2006). The weak coherence account: detail-focused cognitive 44 style in autism spectrum disorders. Journal of Autism and Developmental 45 Disorders, 36(1), 5–25. Hartmann, E.E., Hitchcox, S.M. and Zemon, V.M.: Development of pattern responses 46 to contrast-increment and -decrement stimuli in infants: a VEP measure of 47 functional subsystems in the visual pathway. In Proceedings of the technical 48 digest on noninvasive assessment of the visual system (Optical Society of America, Washington, D.C.), 1992, vol. 1, pp. 146–149. 49 Jasper, H. H. (1958). Appendix to report to Committee on Clinical Examination in 50 EEG: the ten-twenty electrode system of the International Federation. Electro51 encephalography and Clinical Neurophysiology, 10, 371–375. 52 Kaiser, M. D., & Shiffrar, M. (2009). The visual perception of motion by observers with autism spectrum disorders: a review and synthesis. Psychonomic Bulletin & 53 Review, 16(5), 761–777. 54 Kaplan, E. (2004). The M, P, and K pathways of the primate visual system. the visual 55 neurosciences Chalupa LM, Werner JS, (Eds.). MIT Press, Cambridge, MA, 481– 493. 56 Kern, J. K., Trivedi, M. H., Garver, C. R., Grannemann, B. D., Andrews, A. A., Savla, J. S., 57 et al. (2006). Autism, 10(5), 480–494. 58 Koh, H. C., Milne, E., & Dobkins, K. (2010). Contrast sensitivity for motion detection 59 and direction discrimination in adolescents with autism spectrum disorders and their siblings. Neuropsychologia, 48, 4046–4056. 60 Lee, B. B., Pokorny, J., Smith, V., Martin, P. R., & Valbergt, A. (1990). Luminance and 61 chromatic modulation sensitivity of macaque ganglion cells and human 62 observers. Journal of the Optical Society of America A, 7(12), 2223–2236. Leekam, S. R., Nieto, C., Libby, S. J., Wing, L., & Gould, J. (2007). Describing the 63 sensory abnormalities of children and adults with autism. Journal Autism 64 Developmental Disorders, 37(5), 894–910. 65 Lord, C., Rutter, M., DiLavore, P., & Risi, S. (1999). Autism diagnostic observation 66 schedule manual. Los Angeles: Western Psychological Services.

Magnussen, S., & Glad, A. (1975). Brightness and darkness enhancement during flicker: perceptual correlates of neuronal B- and D-systems in human vision. Experimental Brain Research, 22, 399–413. McCleery, J. P., Allman, E. A., Carver, L. J., & Dobkins, K. R. (2007). Abnormal magnocellular pathway visual processing in infants at risk for autism. Biological Psychiatry, 62(9), 1007–1014. McPartland, J., Coffman, M., & Pelphrey, K. (2011). Recent advances in understanding the neural bases of autism spectrum disorder. Current Opinions in Pediatrics, 23(6), 628–632. Merzenich, M. M., Saunders, G., & Tallal, P. (1999). Origins of language impairments for – and impacts of training on – pervasively developmentally disabled children. In: S. Broman, & J. Fletcher (Eds.), Role of neuroplasticity in rare developmental disorders (pp. 365–385). Cambridge, MA: MIT Press. Merzenich, M. M., Tallal, P., Peterson, B., Miller, S. L., & Jenkins, W. M. (1998). Some neurological principles relevant to the origins of – and the cortical plasticity Q16 based remediation of – language learning impairments. In: J. Grafman, & Y. Cristen (Eds.), Neuroplasticity: building a bridge from the laboratory to the clinic (pp. 169–187). New York: Springer-Verlag. Milne, E. (2011). Increased intra-participant variability in children with autistic spectrum disorders: evidence from single-trial analysis of evoked EEG. Frontiers in Psychology, 2, 51. Milne, E., Swettenham, J., Hansen, P., Campbell, R., Jeffries, H., & Plaisted, K. (2002). High motion coherence thresholds in children with autism. Journal of Child Psychology and Psychiatry, 43(2), 255–263. Minshew, N. J., Goldstein, G., & Siegel, D. J. (1997). Neuropsychologic functioning in autism: profile of a complex information processing disorder. Journal of the International Neuropsychological Society, 3(4), 303–316. Mottron, L., Burack, J. A., Iarocci, G., Belleville, S., & Enns, J. T. (2003). Locally oriented perception with intact global processing among adolescents with high-functioning autism: evidence from multiple paradigms. Journal of Child Psychology and Psychiatry, 44(6), 904–913. Mottron, L., Dawson, M., Soulieres, I., Hubert, B., & Burack, J. (2006). Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. Journal of Autism and Developmental Disorders, 36(1), 27–43. Mullen, E. (1995). Mullen scales of early learning. Circle Pines, MN: American Guidance Service, Inc. O'Neill, M., & Jones, R. S. (1997). Sensory-perceptual abnormalities in autism: a case for more research? Journal of Autism and Developmental Disorders, 27(3), 283–293. Pokorny, J., & Smith, V. (1997). Psychophysical signatures associated with magnocellular and parvocellular pathway contrast gain. Journal of the Optical Society of America A, 14, 2477–2486. Rakic, P., Bourgeois, J. P., Eckenhoff, M. F., Zecevic, N., & Goldman-Rakic, P. S. (1986). Concurrent overproduction of synapses in diverse regions of the primate cerebral cortex. Science, 232(4747), 232–235. Roid, G. H. (2003). Stanford-Binet intelligence scales, fifth edition. Itasca, IL: Riverside Publishing. Rubenstein, J. L., & Merzenich, M. M. (2003). Model of autism: increased ratio of excitation/inhibition in key neural systems. Genes Brain and Behavior, 2, 255–267. Rutter, M., LeCouteur, A., & Lord, C (2003). Autism diagnostic interview-revised. Los Angeles: Western Psychological Services. Schiller, P. H. (1982). Central connections of the retinal ON and OFF pathways. Nature, 297(5867), 580–583. Schiller, P. H., Sandell, J. H., & Maunsell, J. H. R. (1986). Functions of the ON and OFF channels of the visual system. Nature, 322, 824–825. Shaw, N. A., & Cant, B. R. (1981). Age-dependent changes in the amplitude of the pattern visual evoked potential. Electroencephalography and Clinical Neurophysiology, 51(6), 671–673. Simmons, D. R., Robertson, A. E., McKay, L. S., Toal, E., McAleer, P., & Pollick, F. E. (2009). Vision in autism spectrum disorders. Vision Research, 49(22), 2705–2739. Spencer, J., O'Brien, J., Riggs, K., Braddick, Atkinson, J., & Wattam-Bell, J. (2000). Motion processing in autism: evidence for a dorsal stream deficiency. Cognitive Neuroscience and Neuropsychology, 11(22), 2765–2767. Sutherland, A., & Crewther, D. P. (2010). Magnocellular visual evoked potential delay with high autism spectrum quotient yields a neural mechanism for altered perception. Brain, 133(P.t 7), 2089–2097. Tootell, R. B., Hamilton, S. L., & Switkes, E. (1988). Functional anatomy of macaque striate cortex. IV. Contrast and magno-parvo streams. The Journal of Neuroscience, 8(5), 1594–1609. Tootell, R. B., Silverman, M. S., Hamilton, S. L., Switkes, E., & De Valois, R. L. (1988). Functional anatomy of macaque striate cortex, 8, 1610–1624The Journal of Q17 Neuroscience, 8, 1610–1624. Vandenbroucke, M. W., Scholte, H. S., van Engeland, H., Lamme, V. A., & Kemner, C. (2008). A neural substrate for atypical low-level visual processing in autism spectrum disorder. Brain, 131, 1013–1024. Victor, J. D., & Mast, J. (1991). A new statistic for steady-state evoked potentials. Electroencephalography and Clinical Neurophysiology, 78(5), 378–388. Wechsler, D. (2003). WISC-IV technical and interpretive manual. San Antonio, TX: Psychological Corporation. Zemon, V., Eisner, W., Gordon, J., Grose-Fifer, J., Tenedios, F., & Shoup, H. (1995). Contrast-dependent responses in the human visual system: childhood through adulthood. International Journal of Neuroscience, 80(1-4), 181–201. Zemon, V., & Gordon, J. (2006). Luminance-contrast mechanisms in humans: visual evoked potentials and a nonlinear model. Vision Research, 46, 4162–4180.

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i

P.M. Weinger et al. / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6

Zemon, V., Gordon, J., & Welch, J. (1988). Asymmetries in ON and OFF visual pathways of humans revealed using contrast-evoked cortical potentials. Visual Neuroscience, 1, 145–150. Zemon, V., Kaplan, E., & Ratliff, F. (1980). Bicuculline enhances a negative component and diminishes a positive component of the visual evoked cortical

9

potential in the cat. Proceedings of the National Academy of Sciences of the United States of America, 77(12), 7476–7478. Zemon, V., Kaplan, E. & Ratliff, F. (1986). In Frontiers of Clinical Neuroscience, v.3: evoked potentials, Cracco, R. Q. & Rodis-Wollner,I. ed., New York: Alan R. Liss. 287–295.

Please cite this article as: Weinger, P. M., et al. Low-contrast response deficits and increased neural noise in children with autism spectrum disorder. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.031i