International Journal of Psychophysiology 96 (2015) 8–15

Contents lists available at ScienceDirect

International Journal of Psychophysiology journal homepage: www.elsevier.com/locate/ijpsycho

Is neural adaptation of the N170 category-specific? Effects of adaptor stimulus duration and interstimulus interval Daniel Feuerriegel a,⁎, Owen F. Churches b, Hannah A.D. Keage a a b

Cognitive Neuroscience Laboratory, University of South Australia, Adelaide, Australia Brain and Cognition Laboratory, Flinders University, Adelaide, Australia

a r t i c l e

i n f o

Article history: Received 21 October 2014 Received in revised form 20 January 2015 Accepted 27 February 2015 Available online 5 March 2015 Keywords: Repetition suppression Neural adaptation P1 N170 P2 Event-related potentials

a b s t r a c t Neural adaptation paradigms have been used in the electrophysiological and neuroimaging literature to characterise neural populations underlying face and object perception. It was recently reported by Nemrodov and Itier (2012) that adaptation of the N170 event-related potential (ERP) component is not stimulus category-specific over rapid adapting stimulus durations (S1 durations) and interstimulus intervals (ISIs). We therefore tested the category-specificity of adaptation over a range of S1 durations and ISIs. Faces and chairs were presented at S1 (for 200, 500 or 1000 ms) and S2 (for 200 ms), over a variable ISI (200 or 500 ms). Mean amplitudes of the P1, N170 and P2 visual ERP components were measured following S1 and S2 stimuli. Faces at S1 led to the smallest (i.e., most adapted) N170 amplitudes to both faces and chairs at S2, more than chairs at S1. N170s at S2 were smallest after a 500 ms S1 duration; but N170 amplitude did not vary over ISI. Effects were also seen for the two surrounding positive components, the P1 and P2. Presenting faces at S1 led to enhanced P1 amplitudes evoked by S2 chair stimuli. The P2 showed the smallest amplitudes following the shorter 200 ms ISI. These results indicate that adaptation of the N170 is not actually category-specific but instead dependent on the S1 category (regardless of S2 category), and may also be influenced by earlier effects at the P1 (i.e., not specific to the N170). This challenges the assumption that N170 category adaptation indexes effects on distinct neural populations that differ between faces and non-face objects. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Human observers are highly proficient at recognising visual objects as abstract categories (e.g., faces, chairs, houses) and individual exemplars within a category (e.g., the face of a friend or one's own house). A long line of research has investigated how category and exemplar information is processed in the visual system. Neuroimaging experiments have identified a distributed network of face and object-selective areas in the ventral temporal cortex (Kanwisher and Yovel, 2006; Haxby et al., 2001; Malach et al., 1995; for a review see Grill-Spector and Weiner, 2014). Event-related potential (ERP) investigations have examined the time course of face and object perception through investigating facesensitive ERP components such as the N170 (Bentin et al., 1996). Researchers have utilised neural adaptation to characterise neuronal populations involved in visual processing (Grill-Spector and Malach, 2001; Koutstaal et al., 2001; Andrews and Ewbank, 2004). Neural adaptation, also called repetition suppression, refers to a reduction in neural population response to repetitions of the same or similar stimuli (for ⁎ Corresponding author at: Cognitive Neuroscience Laboratory, School of Psychology, Social Work and Social Policy, University of South Australia, St Bernards Road, Adelaide, Australia. Tel.: +61 8 830 29939. E-mail address: [email protected] (D. Feuerriegel).

http://dx.doi.org/10.1016/j.ijpsycho.2015.02.030 0167-8760/© 2015 Elsevier B.V. All rights reserved.

reviews see Grill-Spector et al., 2006; Henson, 2003). In the commonly used Adaptor-level Manipulation paradigm, an adapting stimulus (S1) and a subsequent test stimulus (S2) are presented within a trial. By presenting the same stimulus at S2 and manipulating the preceding S1 stimulus along a dimension of interest (e.g., stimulus category), one can assess whether differences along that dimension cause varying degrees of neural adaptation at S2; implying sensitivity to that dimension in the measured neuronal population. For example, if different stimulus categories at S1 cause varying degrees of neural response reduction at S2, this implies sensitivity to stimulus category in the neuronal population of interest. Investigations of the N170 have varied stimulus categories at S1 and S2 to test for stimulus category-specific adaptation. The N170 is thought to index multiple sources of neural activity, including structural encoding of faces and objects, and integration of visual features into a meaningful percept (Bentin et al., 1996; Eimer, 2000; Jacques and Rossion, 2009; Rossion et al., 2000). Since the findings of categoryspecific adaptation for face and hand stimuli by Kovács et al. (2006), most N170 adaptation experiments have presented face and nonface object categories at S1, followed by faces at S2. In these experiments face-specific adaptation is defined as smaller N170s when S2 faces are preceded by S1 faces, compared to other S1 categories. Such an approach relies on the assumption that if stimulus categories other than

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

faces were presented at S2 they would be maximally adapted by S1 stimuli of the same category (Nemrodov and Itier, 2012). This approach also assumes adaptation of distinct neural populations that differ between faces and non-face objects. Following the study of Kovács et al. (2006) others have not reported category-specific adaptation for non-face objects at S2. Recent evidence from rapid adaptation experiments instead suggests that N170 adaptation is not category-specific. A series of experiments by Nemrodov and Itier (2011, 2012) presented face and non-face object categories (houses, cars and chairs) at S1 and S2. They found that the extent of N170 amplitude reduction at S2 depended on the stimulus category at S1. Importantly, these S1 category effects did not interact with S2 category, and therefore were not category-specific (for similar results see Eimer et al., 2010, 2011). Effects specific to S1 category, but not S2 category, challenge assumptions of adaptation of distinct neural populations for different stimulus categories. However, it is unclear whether this S1 categorydependent effect is particular to the rapid 200 ms S1 presentation durations and 200–250 ms ISIs used in experiments that reported these results (Nemrodov and Itier, 2011, 2012; Eimer et al., 2010, 2011). Whether S1 duration and ISI affect the specificity of category adaptation is unclear, as these experimental factors have not been systematically studied. In addition to the N170, the P1 and P2 visual ERP components also show stimulus repetition effects to faces and non-face objects. P1 amplitude has been shown to differ by S1 category, and is typically enhanced for S2 stimuli following S1 faces (e.g., Kovács et al., 2006; Nemrodov and Itier, 2012; Xu et al., 2012). Adaptation during the time range of the P2 is sensitive to face identity (Schweinberger et al., 2004, 2002; Xu et al., 2012) although identity-specific N170 adaptation has also been reported following long (3-second) S1 durations (Jacques et al., 2007; Caharel et al., 2009, 2011). The obvious step to determine whether category-specific adaption does operate, as captured by ERPs, is to systematically vary S1 and ISI durations. Varying S1 duration will identify whether adaptation is influenced by prolonged stimulus processing, as is the case in the auditory system (Lanting et al., 2013); while varying ISI will index the speed of recovery from adaptation. Therefore we systematically tested for differences in adaptation across S1 duration and ISI following faces and nonface objects. As the neural mechanisms of adaptation are currently unresolved (Grill-Spector et al., 2006; Gotts et al., 2012) identifying conditions under which category-specific adaptation occurs is necessary to validate category adaptation paradigms. In addition to the N170 we measured the P1 and P2 visual ERP components to assess whether category adaptation effects are specific to the N170. We expected to find S1 category-dependent effects over rapid (200 ms) S1 and ISI durations, and category-specific N170 adaptation following longer (500 ms and 1000 ms) S1 durations. 2. Materials and methods 2.1. Participants Twenty people participated in this experiment. One participant was excluded from analyses due to poor task performance (correct responses to b70% of target trials) and 3 others were excluded due to excessive EEG artefacts. The remaining 16 participants (6 males) were 20–35 years old (mean age 25.5 ± 4.6 years). All participants had normal or corrected-to-normal vision. Fifteen were right-handed and 1 was left-handed as assessed by the Flinders Handedness Survey (Nicholls et al., 2013). This study was approved by the ethics committee of the University of South Australia. 2.2. Stimuli Greyscale images of faces and chairs were used. 50 face images (25 male) of neutral expression were taken from the Karolinska Directed

9

Emotional Faces (Lundqvist et al., 1998). Frontal photographs of 50 chairs were taken under uniform lighting conditions. Chair photograph backgrounds were changed to white using Adobe Photoshop (www. adobe.com). Face and chair images were cropped and resized so that at a distance of 60 cm S1 stimuli subtended approximately 4.3 × 4.8° of visual angle. S2 stimuli were created 5% larger than adaptor stimuli to minimise retinal persistence effects. The SHINE toolbox (Willenbockel et al., 2010, www.mapageweb.umontreal.ca/ gosselif/shine) was used to match mean luminance and contrast across all images. Mean normalised grayscale pixel intensity was 0.44 and mean RMS contrast was 0.19. Stimuli were presented against a black background. Examples of stimuli are shown in Fig. 1. Targets were created from the resulting stimuli by adding a red rectangle to S1 and S2 images (as done by Eimer et al., 2010). Target stimuli were equally allocated across different experimental conditions, S1 and S2 presentation and face and chair images. There were equal numbers of male and female face targets. 2.3. Procedure Participants sat in a sound-attenuated well-lit testing room 60 cm in front of a 474 × 296 mm LED computer monitor. Stimuli were presented using STIM2 presentation software (Compumedics). Behavioural responses were recorded in STIM2 using a four-button response box. In each trial, S1 and S2 stimuli were presented separated by an ISI. Face and chair stimuli were presented at S1 and S2. All stimulus combinations were presented in six different S1 duration and ISI combinations: S1 durations of 200 ms, 500 ms and 1000 ms and ISI durations of 200 ms and 500 ms. S2 duration remained constant at 200 ms. The inter-trial interval (ITI) was 1200 ms. An example trial is shown in Fig. 1. This study included a total of 24 non-target conditions (4 stimulus combinations × 3 S1 durations × 2 ISI durations). Equal numbers of trials of each condition were pseudo-randomly interleaved into 8 blocks of 264 trials (11 min). Each block included 24 target trials. Block order was counterbalanced across participants. Trial order and stimulus image order were pseudo-randomised so that the same chair or face image would not occur within the same trial, and that the S2 stimulus image of one trial would not be the S1 stimulus image in the next trial, and that no more than four faces or four chairs would appear consecutively. Equal numbers of chairs and faces at S1, S2 and overall were included in each block and across the experiment. Equal numbers of trials of each S1 duration and ISI combination condition were included within each block. Over the entire study run each face and chair image was presented in pseudorandom order around 42 times. 2.4. Experimental task Participants were instructed to press the left or right button on a response pad upon seeing the S2 stimulus if either stimulus in a trial contained a red rectangle. Participants were instructed to respond as quickly and accurately as possible. Response buttons were counterbalanced across participants. Responses within 1300 ms from S2 stimulus onset were counted as correct responses. A short practice session was completed before starting the experiment. Targets occurred in 9.1% of trials. Target trials and trials with button responses were excluded from ERP analyses. Of the 2112 trials presented during the experiment 1920 trials did not include target stimuli (80 trials per condition for 24 conditions). 2.5. EEG recording and data processing The electroencephalogram (EEG) was recorded using 64 Ag/AgCl electrodes mounted in a Quick-Cap (Compumedics) according to the extended 10–20 international system using a nose reference and a ground located between sites Fz and FPz. Six additional electrodes

10

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

Fig. 1. Example trial and stimulus combinations. a In each trial the S1 stimulus was presented for 200 ms, 500 ms or 1000 ms. The ISI duration varied between 200 ms or 500 ms. b Faces and chairs were presented at S1 and S2 in all stimulus combinations. The example trial shown here contains a target stimulus (red rectangle) at S2. Coloured lines between S1 and S2 stimuli in each stimulus combination denote colours used for these stimulus combinations in the grand average S2 waveforms in Fig. 3.

were added to the standard montage: two electrodes recorded horizontal eye movements and were placed 1 cm from the outer canthi of each eye, and two electrodes recorded vertical eye movements and were placed above and below the left eye, and two electrodes on the left and right mastoids. EEG was amplified using a Synamps amplifier (Compumedics Neuroscan) using a band-pass filter of 0.05–200 Hz and digitised using SCAN 4.5 acquisition software (Compumedics) at a sample rate of 1000 Hz. Electrode impedances were kept below 5 kΩ. EEG data were processed using EEGLab (Delorme and Makeig, 2004) and ERPLab (Lopez-Calderon and Luck, 2014). Four channels were removed from the dataset: HEO, VEO, CB1 and CB2. Channels containing excessive artefacts were replaced using spline interpolation for each participant. EEG was epoched relative to a 100 ms pre-S1 baseline until 400 ms after S2 onset. Data was then re-referenced to an average reference and low-pass filtered at 30 Hz using an IIR Butterworth filter (12 dB/octave). Epochs containing artefacts (deviations of ± 100 μV from baseline) in any channel were then rejected. This method ensures exact correspondence between the number of trials kept in the average of the S1 stimulus and the number of trials kept in the average of the S2 stimulus preceded by this S1 stimulus (Nemrodov and Itier, 2012; Eimer et al., 2010). EEG waveforms were averaged separately for S1 and S2 in each of the 24 conditions. Responses to S1 stimuli were averaged relative to a 100 ms pre-stimulus baseline and responses to S2 stimuli were averaged relative to a baseline from 50 ms before until 50 ms after S2 stimulus onset (as done by Nemrodov and Itier, 2011, 2012; Eimer et al., 2010). Late activity from S1-evoked ERPs overlapped with S2 waveforms when the ISI was 200 ms. For each combination of S1 category and S1 duration the ERP waveform of the ISI period in the 500 ms ISI condition (which contains late activity from S1) was subtracted from the 200 ms prestimulus to 300 ms post-stimulus period of the equivalent ISI 200 ms condition (which contains the late activity from S1 plus the S2-evoked waveform over the same time period relative to S1 onset). This method isolates the S2 waveform from late S1-evoked activity (as done by Lanting et al., 2013; Thorpe et al., 1996).

Mean amplitudes of the N170 were measured from lateral posterior electrodes P7 and P8 where N170 amplitudes were maximal to both faces and chairs (as done by Eimer et al., 2010; Kovács et al., 2006). P1 and P2 amplitudes were measured from PO7 and PO8 where amplitudes were maximal for both components. S2 waveforms were slightly delayed relative to S1. Accordingly, P1 mean amplitudes were calculated as the average of activity between 120–160 ms from S1 onset, and from 125–165 ms from S2 onset. N170 mean amplitudes were calculated as the average of activity between 160–210 ms from S1 onset, and from 170–220 ms from S2 onset. P2 mean amplitudes were calculated as the average of activity between 220–300 ms from S1 onset, and from 230–310 ms from S2 onset. 2.6. Statistical analysis Five-way repeated measures analyses of variance (ANOVAs) were performed on S1 and S2-evoked P1, N170 and P2 mean amplitudes with factors S1 category (face, chair), S2 category (face, chair), S1 duration (200 ms, 500 ms, 1000 ms), ISI duration (200 ms, 500 ms) and hemisphere (left — P7/PO7, right — P8/PO8). No main effects or interactions involving hemisphere were identified for P1 or N170 component amplitudes, and so this factor was collapsed to improve power in analyses of these components. Greenhouse–Geisser corrections to degrees of freedom were performed when appropriate. Paired samples t-tests were conducted to investigate significant main effects and interactions using a planned contrast method where only comparisons relevant to the hypotheses were run. The Benjamini– Hochberg procedure (Benjamini and Hochberg, 1995) was used to control experimentwise and familywise false discovery rates. 3. Results 3.1. Task performance Overall task performance was 92.2% ± 6.5%. Mean response time was 482 ms ± 77 ms.

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

3.2. S1 stimulus analyses Grand average waveforms to face and chair S1 stimuli are displayed in Fig. 2. For P1 mean amplitudes no main effects or interactions reached alpha-corrected significance. N170s evoked by faces were larger (more negative) than those evoked by chairs [main effect of S1 category, F(1,15) = 20.35, p b .001]. Analyses of P2 mean amplitudes revealed a main effect of hemisphere, showing larger P2 amplitudes in the right hemisphere [F(1,15) = 19.56, p = .001]. 3.3. S2 stimulus analyses 3.3.1. P1 ERP component Grand average waveforms to S2 stimuli are displayed in Fig. 3. Analyses of S2-evoked P1 amplitudes found 3-way interactions of S1 duration by S1 category by S2 category [F(2,30) = 8.25, p = .001] and ISI by S1 duration by S2 category [F(2,30) = 13.88, p b .001]. The S1 duration by S1 category by S2 category interaction revealed that presenting faces at S1 (compared to chairs at S1) led to larger P1 amplitudes evoked by chairs at S2, when the S1 duration was 200 ms (p = .003) and 1000 ms (p = .002). No differences by S1 category were found for S2 faces. Further analyses of the ISI by S1 duration by S2 category interaction compared across S1 durations for each S2 category and ISI combination. Following 200 ms ISIs, P1 amplitudes evoked by chairs at S2 were larger after S1 durations of 200 ms compared to 500 ms (p = .003) and 1000 ms (p = .001). No differences across S1 duration or ISI were found for S2 faces. 3.3.2. N170 ERP component Analyses of S2-evoked N170 amplitudes yielded a main effect of S2 category showing that faces at S2 evoked larger N170s than chairs at S2 [F(1,15) = 20.50, p b .001]. In addition, S2 stimuli preceded by S1 faces evoked smaller N170 amplitudes than when preceded by S1 chairs [main effect of S1 category, F(1,15) = 42.68, p b .001]. There was a trend towards a main effect of S1 duration, but this did not reach alphacorrected significance (p = .099). 3.3.3. P1–N170 difference scores P1–N170 mean amplitude difference scores at P7 and P8 were calculated to assess N170 amplitude effects corrected for differences in P1 amplitudes (as done by Kuefner et al., 2010; Peykarjou et al., 2014). There were no main effects or interactions involving hemisphere, and so this factor was collapsed to improve power. The main effect of S1 category [F(1,15) = 15.03, p = .001] was replicated using this measure. The main effect of S1 duration reached significance [F(2,30) = 9.76,

11

p = .001] with smaller P1–N170 difference scores when the ISI was 500 ms compared to both 200 ms (p = .002) and 1000 ms (p b .001).

3.3.4. P2 ERP component S2-evoked P2 amplitudes were larger over the right hemisphere [main effect of hemisphere, F(1,15) = 19.11, p = .001]. There was a main effect of ISI showing smaller P2 amplitudes following 200 ms compared to 500 ms ISIs [F(1,15) = 77.47, p b .001]. Faces at S2 evoked larger P2 amplitudes than chairs at S2 [main effect of S2 category, F(1,15) = 14.89, p = .002]. P2 amplitudes were also larger when S2 stimuli were preceded by S1 faces compared to S1 chairs [main effect of S1 category, F(1,15) = 12.54, p = .003]. In addition, there were interactions of ISI by S1 duration [F(2,30) = 5.29, p = .011], ISI by S1 category [F(1,15) = 10.02, p = .006] and S1 category by S2 category [F(1,15) = 22.75, p b .001]. There were interactions of hemisphere by ISI [F(1,15) = 29.20, p b .001] and ISI by S1 category by S2 category interaction [F(1,15) = 16.53, p = .001], however follow-up tests found no interaction effects that reached alpha-corrected significance. Analyses of the ISI by S1 duration interaction revealed S1 duration effects on P2 amplitudes, with smaller P2 amplitudes following 500 ms compared to 200 ms S1 durations when the ISI was 200 ms (p = .007) and smaller P2 amplitudes following 1000 ms compared to 500 ms S1 durations after ISIs of 500 ms (p b .001). The S1 category by S2 category interaction revealed that presenting faces at S1 (compared to chairs at S1) led to increased P2 amplitudes evoked by chairs at S2 (p b .001), however no differences by S1 category were found for S2 faces.

4. Discussion We demonstrated the dynamic nature of neural adaptation by testing for stimulus category-specific effects while systematically varying S1 duration and ISI. Our N170 findings indicate that adaptation in vision is not category-specific over the tested S1 durations and ISIs; as faces at S1 adapted responses to both chairs and faces at S2. This also suggests that the category-specificity of adaptation is critically dependent on the S1 duration, and may be category-specific only after long (5-s) S1 durations as used in Kovács et al. (2006). The observed S1 categorydependent P1 amplitude enhancements imply that category adaptation effects begin earlier than the N170. Effects of S1 duration and ISI on adaptation were different for the P1, N170 and P2, suggesting that neural populations indexed by these components adapt over different timescales. The implications of our results challenge long held assumptions regarding the category-specificity and timing of neural adaptation in vision.

Fig. 2. a Grand average waveforms to S1 stimuli. Grand averages were derived from averaging across electrodes P7/8 and PO7/8. b Scalp maps of the average amplitude evoked by S1 stimuli during the N170 time range (160–210 ms).

12

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

Fig. 3. Grand average waveforms to S2 stimuli. Grand averages were derived from averaging across electrodes P7/8 and PO7/8. a Waveforms plotted by S1 duration (left), ISI (centre) and stimulus category (right). b Waveforms for all stimulus combinations following each ISI and S1 duration. c Scalp maps of mean amplitude during the time range of the N170 (170–220 ms) for each condition.

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

13

4.1. Effects of S1 duration

4.3. Effects of stimulus category

We found strongest adaptation of the N170 following S1 durations of 500 ms (as compared to 200 and 1000 ms), indicating that stimulus category adaptation does not simply increase as a function of S1 duration. Similar to our study, Zago et al. (2005) compared across S1 durations of 40–1900 ms using fMRI, and reported strongest adaptation of visual and frontal neurons following S1 durations of 250 ms. They interpreted these results as the visual system developing an exhaustive representation of stimulus features over continued exposure up to 250 ms. With longer exposure the representation was proposed to be reduced to features essential for individual object recognition. This reduction would also lead to a decrease in neurons showing adaptation (see also Wiggs and Martin, 1998). An alternative proposal is that neural populations that process different stimulus characteristics adapt over different S1 durations. As adaptation in our study was not category-specific, S1 exposure up to 500 ms may have adapted neural populations common to faces and chairs. In contrast, longer S1 durations may preferentially adapt neural populations that encode other stimulus features, rather than the neural populations that exhibited adaptation effects in our study. These features may include viewpoint, spatial position and facial identity, which have been found to adapt after 5-s (but not 500 ms or shorter) S1 durations (Fang et al., 2007; Kovács et al., 2007). This could be further investigated by testing S1 duration effects on adaptation of different types of stimulus information.

Faces at S1 preferentially reduced N170s evoked by both S2 faces and chairs over all S1 durations and ISIs. We followed a standard procedure of selecting electrodes where the N170 was maximal (P7/P8; for both chairs and faces), however, to ensure that effects were not a product of this selection, we ran another model with a larger array of electrodes (P7/8, P5/6, PO7/8, O1/2) and found that the results were consistent. These results indicate that N170 adaptation is not category-specific or face-specific as previously thought, but instead dependent on the category of the S1 stimulus. Another likely interpretation is that category adaptation over the timescales in our experiment (which are representative of a large number of category adaptation studies) qualitatively differs from category-specific adaptation induced by long (5-s) S1 durations as in Kovács et al. (2006). The limited range of S1 durations in our study prevented us from comparing differences between long and short adaptation. Our results therefore may not be representative of long S1 duration adaptation studies, and suggest that future studies testing for category-specific effects should use multiple second S1 durations. These results also suggest that over the tested timescales category adaptation does not affect distinct neural populations for faces and non-face objects. Rather, a large proportion of adapted neurons appear to respond to both stimulus categories. These neurons may be tuned to certain features that differ by stimulus category (e.g., Desimone, 1984). If so, then differences in firing rate by stimulus category at S1 could modulate the extent of adaptation at S2, leading to S1 categorydependent effects (Nemrodov and Itier, 2012). Such a mechanism would be different from adaptation to repeated images, in which reductions in neural firing are specific to image repetition, regardless of neuronal firing rate at S1 (Sawamura et al., 2006). As we presented only faces and chairs at S1 and S2, we could not determine the stimulus properties that affect S1 category-dependent adaptation. If category adaptation is determined by the tuning properties of neurons then identifying such properties would allow characterisation of neuronal populations that are not stimulus category-selective. Neuroimaging paradigms could also investigate the anatomical distribution of S1 category-dependent adaptation to determine the locus of the effects observed in the current study. Our experiment differs in several ways from Kovács et al. (2006) that reported category-specific N170 adaptation. Their experiment employed longer (5-s) S1 durations and presented gender-morphed face and hand images at S2. Following 5-second S1 durations N170 adaptation is sensitive to facial identity (Kovács et al., 2007). It is possible that gender or identity-related shape information in the morphed images overlapped between S1 and S2 in Kovács et al. (2006) and that this visual information was also adapted at the N170. Accordingly, category adaptation in their study may have also indexed contributions from N170 identity adaptation. Alternatively, the S1 durations in our study may have been too short to observe category-specific adaptation, given that longer S1 durations have been shown to increase the specificity of adaptation (Fang et al., 2007). This remains to be tested using a range of face and non-face object stimuli over long (5-s) S1 durations. Analyses of the P1 show that S1 category-dependent effects may not be restricted to the N170. Faces at S1 led to enhanced P1 amplitudes evoked by S2 chairs. Similar S1 category-dependent P1 effects can be observed in the grand average waveforms of other studies (Kovács et al., 2006, 2007; Schweinberger et al., 2007; Kloth et al., 2010; Nemrodov and Itier, 2012; Schinkel et al., 2014; Xu et al., 2012). These P1 enhancements raise the possibility that category adaptation at the N170 is affected by feed-forward adaptation effects from earlier visual processing (Kohn, 2007; Tolias et al., 2005) which can be problematic for localising adaptation effects to the N170 (Kovács et al., 2006). The current study could not determine whether the observed P1 and N170 category effects can be dissociated, as face and chair stimuli also differed with regard to visual properties such as Fourier amplitude

4.2. Effects of ISI There were no differences in the extent of N170 adaptation across 200 ms and 500 ms ISI durations, suggesting that this adaptation does not recover rapidly over the ISI (see also Kuehl et al., 2013). Our results differ from the MEG study of Harris and Nakayama (2007) that tested ISIs of 100–600 ms and found increasing recovery from adaptation with longer ISI duration. As Harris and Nakayama (2007) tested over short SOAs (300–800 ms) it is likely that late S1-evoked activity influenced S2 baseline and poststimulus event-related fields. Importantly, these effects would differ by ISI duration. As we tested across multiple S1 durations our ISI results are unlikely to be caused by late S1-evoked potentials on S2 waveforms. However, it is likely that the range of ISI durations in our study was not broad enough to detect effects on adaptation. N170 category adaptation does appear to recover over longer ISIs, as studies using N 1.2 s ISIs generally do not report category adaptation effects (Schweinberger et al., 2002; Engst et al., 2006; Neumann and Schweinberger, 2008). The P2 component displayed more rapid recovery from adaptation over the ISI period, displaying similar amplitudes to unadapted S1 stimuli after 500 ms ISIs. These category adaptation effects in our experiment appear to be distinct from P2 adaptation to repeats of the same stimulus image, which has been found following ISIs up to 4800 ms (Gruber and Muller, 2005; Henson et al., 2004; Walther et al., 2013; Xu et al., 2012). It is unclear if different rates of recovery for the N170 and P2 relate to behavioural or perceptual effects. This possibility should be taken into account when comparing across experiments that differ in respect to S1 duration or ISI. When interpreting these S1 duration and ISI results it is important to note that our experiment did not control for differences in attention across S1 duration and ISI. Such differences may explain the observed P1 amplitude differences across S1 duration that were not consistent across ISIs. Future research could control for attention effects on S2-evoked ERPs by comparing the extent of stimulusspecific adaptation (i.e., component amplitude differences between the same and different stimuli at S1 and S2) following each S1 duration or ISI.

14

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15

spectra and stimulus shape. Experiments using blocked designs (Maurer et al., 2008; Mercure et al., 2011) or trials with two adapting stimuli (Amihai et al., 2011) did not report the N170 or P1 effects observed in our study. Amihai et al. (2011) presented faces and Easter eggs with the same global shape, suggesting that shape information contributes to P1 and N170 category effects. The other two studies included ISIs of 1 s or longer (Maurer et al., 2008; Mercure et al., 2011) which may too long to find category effects in event-related designs (Schweinberger et al., 2002; Engst et al., 2006; Neumann and Schweinberger, 2008). Until the conditions that support category-specific adaptation can be identified previous reports of face-specific adaptation should be interpreted with caution. Future adaptation designs should be validated by presenting control stimuli at both S1 and S2 to ensure specificity of adaptation for the tested stimulus dimension. 4.4. Conclusions Our results show that S1 duration and ISI have distinct effects on neural adaptation at different stages of visual processing. The stimulus category effects observed in the present study suggest that category adaptation is S1 duration-dependent and is not category-specific over short timescales, and that many previous reports of face-specific adaptation can be attributed to S1 category-dependent effects. These results question the validity of widely used category adaptation paradigms to characterise neuronal populations involved in visual processing of faces and non-face objects. Acknowledgements We thank Ms. Lisa Kurylowicz for assistance with data collection. HADK was funded by a NHMRC Early Career Fellowship (GNT568890). References Amihai, I., Deouell, L.Y., Bentin, S., 2011. Neural adaptation is related to face repetition irrespective of identity: a reappraisal of the N170 effect. Exp. Brain Res. 209, 193–204. Andrews, T.J., Ewbank, M.P., 2004. Distinct representations for facial identity and changeable aspects of faces in the human temporal lobe. Neuroimage 23, 905–913. Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Methodol. 289–300. Bentin, S., Allison, T., Puce, A., Perez, E., McCarthy, G., 1996. Electrophysiological studies of face perception in humans. J. Cogn. Neurosci. 8, 551–565. Caharel, S., d'Arripe, O., Ramon, M., Jacques, C., Rossion, B., 2009. Early adaptation to repeated unfamiliar faces across viewpoint changes in the right hemisphere: evidence from the N170 ERP component. Neuropsychologia 47, 639–643. Caharel, S., Jacques, C., d'Arripe, O., Ramon, M., Rossion, B., 2011. Early electrophysiological correlates of adaptation to personally familiar and unfamiliar faces across viewpoint changes. Brain Res. 1387, 85–98. Delorme, A., Makeig, S., 2004. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J. Neurosci. Methods 134, 9–21. Desimone, R., Albright, T.D., Gross, C.G., Bruce, C., 1984. Stimulus-selective properties of inferior temporal neurons in the macaque. J. Neurosci. 4, 2051–2062. Eimer, M., 2000. The face-specific N170 component reflects late stages in the structural encoding of faces. Neuroreport 11, 2319–2324. Eimer, M., Kiss, M., Nicholas, S., 2010. Response profile of the face-sensitive N170 component: a rapid adaptation study. Cereb. Cortex 20, 2442–2452. Eimer, M., Gosling, A., Nicholas, S., Kiss, M., 2011. The N170 component and its links to configural face processing: a rapid neural adaptation study. Brain Res. 1376, 76–87. Engst, F.M., Martin-Loeches, M., Sommer, W., 2006. Memory systems for structural and semantic knowledge of faces and buildings. Brain Res. 1124, 70–80. Fang, F., Murray, S.O., He, S., 2007. Duration-dependent FMRI adaptation and distributed viewer-centered face representation in human visual cortex. Cereb. Cortex 17, 1402–1411. Gotts, S.J., Chow, C.C., Martin, A., 2012. Repetition priming and repetition suppression: a case for enhanced efficiency through neural synchronization. Cogn. Neurosci. 3, 227–237. Grill-Spector, K., Malach, R., 2001. fMR-adaptation: a tool for studying the functional properties of human cortical neurons. Acta Psychol. 107, 293–321. Grill-Spector, K., Weiner, K.S., 2014. The functional architecture of the ventral temporal cortex and its role in categorization. Nat. Rev. Neurosci. 15, 536–538. Grill-Spector, K., Henson, R., Martin, A., 2006. Repetition and the brain: neural models of stimulus-specific effects. Trends Cogn. Sci. 10, 14–23.

Gruber, T., Muller, M.M., 2005. Oscillatory brain activity dissociates between associative stimulus content in a repetition priming task in the human EEG. Cereb. Cortex 15, 109–116. Harris, A., Nakayama, K., 2007. Rapid face-selective adaptation of an early extrastriate component in MEG. Cereb. Cortex 17, 63–70. Haxby, J.V., Gobbini, M.I., Furey, M.L., Ishai, A., Schouten, J.L., Pietrini, P., 2001. Distributed and overlapping representations of faces and objects in ventral temporal cortex. Science 293, 2425–2430. Henson, R.N., 2003. Neuroimaging studies of priming. Prog. Neurobiol. 70, 53–81. Henson, R.N., Rylands, A., Ross, E., Vuilleumeir, P., Rugg, M.D., 2004. The effect of repetition lag on electrophysiological and haemodynamic correlates of visual object priming. Neuroimage 21, 1674–1689. Jacques, C., Rossion, B., 2009. The initial representation of individual faces in the right occipito-temporal cortex is holistic: electrophysiological evidence from the composite face illusion. J. Vis. 9 (8), 1–16. Jacques, C., d'Arripe, O., Rossion, B., 2007. The time course of the inversion effect during individual face discrimination. J. Vis. 7, 3. Kanwisher, N., Yovel, G., 2006. The fusiform face area: a cortical region specialized for the perception of faces. Philos. Trans. R. Soc. Lond. B Biol. Sci. 361, 2109–2128. Kloth, N., Schweinberger, S.R., Kovács, G., 2010. Neural correlates of generic versus gender-specific face adaptation. J. Cogn. Neurosci. 22, 2345–2356. Kohn, A., 2007. Visual adaptation: physiology, mechanisms, and functional benefits. J. Neurophysiol. 97, 3155–3164. Koutstaal, W., Wagner, A.D., Rotte, M., Maril, A., Buckner, R.L., Schacter, D.L., 2001. Perceptual specificity in visual object priming: Functional magnetic resonance imaging evidence for a laterality difference in fusiform cortex. Neuropsychologia 39, 184–199. Kovács, G., Zimmer, M., Bankó, É., Harza, I., Antal, A., Vidnyánszky, Z., 2006. Electrophysiological correlates of visual adaptation to faces and body parts in humans. Cereb. Cortex 16, 742–753. Kovács, G., Zimmer, M., Harza, I., Vidnyánszky, Z., 2007. Adaptation duration affects the spatial selectivity of facial aftereffects. Vis. Res. 47, 3141–3149. Kuefner, D., de Heering, A., Jacques, C., Palmero-Soler, E., Rossion, B., 2010. Early visually evoked electrophysiological responses over the human brain (P1, N170) show stable patterns of face-sensitivity from 4 years to adulthood. Front. Hum. Neurosci. 3, 67. Kuehl, L.K., Brandt, E.S., Hahn, E., Dettling, M., Neuhaus, A.H., 2013. Exploring the time course of N170 repetition suppression: a preliminary study. Int. J. Psychophysiol. 87, 183–188. Lanting, C.P., Briley, P.M., Sumner, C.J., Krumbholz, K., 2013. Mechanisms of adaptation in human auditory cortex. J. Neurophysiol. 110, 973–983. Lopez-Calderon, J., Luck, S.J., 2014. ERPLAB: an open-source toolbox for the analysis of event-related potentials. Front. Hum. Neurosci. 8, 213. Lundqvist, D., Flykt, A., Öhman, A., 1998. The Karolinska Directed Emotional Faces. Karolinska Institute, Stockholm, Sweden. Malach, R., Reppas, J.B., Benson, R.R., Kwong, K.K., Jiang, H., Kennedy, W.A., Ledden, P.J., Brady, T.J., Rosen, B.R., Tootell, R.B., 1995. Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. Proc. Natl. Acad. Sci. U. S. A. 92, 8135–8139. Maurer, U., Rossion, B., McCandliss, B.D., 2008. Category specificity in early perception: face and word N170 responses differ in both lateralization and habituation properties. Front. Hum. Neurosci. 2, 18. Mercure, E., Kadosh, K.C., Johnson, M.H., 2011. The N170 shows differential repetition effects for faces, objects, and orthographic stimuli. Front. Hum. Neurosci. 5. Nemrodov, D., Itier, R.J., 2011. The role of eyes in early face processing: a rapid adaptation study of the inversion effect. Br. J. Psychol. 102, 783–798. Nemrodov, D., Itier, R.J., 2012. Is the rapid adaptation paradigm too rapid? Implications for face and object processing. Neuroimage 61, 812–822. Neumann, M.F., Schweinberger, S.R., 2008. N250r and N400 ERP correlates of immediate famous face repetition are independent of perceptual load. Brain Res. 1239, 181–190. Nicholls, M.E., Thomas, N.A., Loetscher, T., Grimshaw, G.M., 2013. The Flinders Handedness survey (FLANDERS): A brief measure of skilled hand preference. Cortex 49, 2914–2926. Peykarjou, S., Pauen, S., Hoehl, S., 2014. How do 9-month-old infants categorize human and ape faces? A rapid repetition ERP study. Psychophysiology 51, 866–878. Rossion, B., Gauthier, I., Tarr, M.J., Despland, P., Bruyer, R., Linotte, S., Crommelinck, M., 2000. The N170 occipito-temporal component is delayed and enhanced to inverted faces but not to inverted objects: an electrophysiological account of face-specific processes in the human brain. NeuroReport 11, 69–74. Sawamura, H., Orban, G.A., Vogels, R., 2006. Selectivity of neuronal adaptation does not match response selectivity: a single-cell study of the FMRI adaptation paradigm. Neuron 49, 307–318. Schinkel, S., Ivanova, G., Kurths, J., Sommer, W., 2014. Modulation of the N170 adaptation profile by higher level factors. Biol. Psychol. 97, 27–34. Schweinberger, S.R., Pickering, E.C., Jentzsch, I., Burton, A.M., Kaufmann, J.M., 2002. Eventrelated brain potential evidence for a response of inferior temporal cortex to familiar face repetitions. Brain Res. Cogn. Brain Res. 14, 398–409. Schweinberger, S.R., Huddy, V., Burton, A.M., 2004. N250r: a face-selective brain response to stimulus repetitions. Neuroreport 15, 1501–1505. Schweinberger, S.R., Kloth, N., Jenkins, R., 2007. Are you looking at me? Neural correlates of gaze adaptation. Neuroreport 18, 693–696. Thorpe, S., Fize, D., Marlot, C., 1996. Speed of processing in the human visual system. Nature 381, 520–522. Tolias, A.S., Keliris, G.A., Smirnakis, S.M., Logothetis, N.K., 2005. Neurons in macaque area V4 acquire directional tuning after adaptation to motion stimuli. Nat. Neurosci. 8, 591–593. Walther, C., Schweinberger, S.R., Kaiser, D., Kovacs, G., 2013. Neural correlates of priming and adaptation in familiar face perception. Cortex 49, 1963–1977.

D. Feuerriegel et al. / International Journal of Psychophysiology 96 (2015) 8–15 Wiggs, C.L., Martin, A., 1998. Properties and mechanisms of perceptual priming. Curr. Opin. Neurobiol. 8, 227–233. Willenbockel, V., Sadr, J., Fiset, D., Horne, G.O., Gosselin, F., Tanaka, J.W., 2010. Controlling low-level image properties: the SHINE toolbox. Behav. Res. Methods 42, 671–684.

15

Xu, M., Lauwereyns, J., Iramina, K., 2012. Dissociation of category versus item priming in face processing: an event-related potential study. Cogn. Neurodyn. 6, 155–167. Zago, L., Fenske, M.J., Aminoff, E., Bar, M., 2005. The rise and fall of priming: how visual exposure shapes cortical representations of objects. Cereb. Cortex 15, 1655–1665.

Is neural adaptation of the N170 category-specific? Effects of adaptor stimulus duration and interstimulus interval.

Neural adaptation paradigms have been used in the electrophysiological and neuroimaging literature to characterise neural populations underlying face ...
1MB Sizes 0 Downloads 5 Views