Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 Neurobiology of Learning and Memory xxx (2015) xxx–xxx 1

Contents lists available at ScienceDirect

Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme 5 6

4

Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding

7

Eva M. Bauch a,⇑, Nico Bunzeck a,b

3

8 9 10 11 1 2 3 5 14 15 16 17 18 19 20 21 22 23 24

a b

Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany Department of Psychology, University of Lübeck, Ratzeburger Allee 160, 23562 Lübeck, Germany

a r t i c l e

i n f o

Article history: Received 31 October 2014 Revised 16 June 2015 Accepted 17 June 2015 Available online xxxx Keywords: Magnetoencephalography Beta power Memory formation Pain anticipation

a b s t r a c t In humans, the temporal and oscillatory dynamics of pain anticipation and its effects on long-term memory are largely unknown. Here, we investigated this open question by using a previously established behavioral paradigm in combination with magnetoencephalography (MEG). Healthy human subjects encoded a series of scene images, which was combined with cues predicting an aversive electric shock with different probabilities (0.2, 0.5 or 0.8). After encoding, memory for the studied images was tested using a remember/know recognition task. Behaviorally, pain anticipation did not modulate recollection-based recognition memory per se, but interacted with the perceived unpleasantness of the electric shock [visual analogue scale rating from 1 (not unpleasant) to 10 (highly unpleasant)]. More precisely, the relationship between pain anticipation and recollection followed an inverted u-shaped function the more unpleasant the shocks were rated by a subject. At the physiological level, this quadratic effect was mimicked in the event-related magnetic fields associated with successful memory formation (‘DM-effect’) 450 ms after image onset at left frontal sensors. Importantly, across all subjects, shock anticipation modulated oscillatory power in the low beta frequency range (13–20 Hz) in a linear fashion at left temporal sensors. Taken together, our findings indicate that beta oscillations provide a generic mechanism underlying pain anticipation; the effect on subsequent long-term memory, on the other hand, is much more variable and depends on the level of individual pain perception. As such, our findings give new and important insights into how aversive motivational states can drive memory formation. Ó 2015 Published by Elsevier Inc.

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

46 47

1. Introduction

48

Rewards and punishments are motivating factors of goal-directed behavior in learning animals and humans. In line with the evolutionary need of a rapid learning system, electrophysiological studies (magnetoencephalography [MEG], electroencephalography [EEG]) show that both, nociceptive information and the anticipation of aversive stimuli, are signaled in various brain regions already at 100 ms after stimulus presentation (Garcia-Larrea, Frot, & Valeriani, 2003; Iannetti, Zambreanu, Cruccu, & Tracey, 2005; Pizzagalli, Greischar, & Davidson, 2003; Ploner, Gross, Timmermann, Pollok, & Schnitzler, 2006a) (aversive anticipation: Dolan, Heinze, Hurlemann, & Hinrichs, 2006; Weymar, Bradley, Hamm, & Lang, 2013). Particularly, recordings in animals show that dopaminergic neurons in the substantia nigra/ventral tegmental area (SN/VTA) rapidly respond (onset latency of 100 ms) not only to cues associated with reward

49 50 51 52 53 54 55 56 57 58 59 60 61 62

⇑ Corresponding author. Fax: +49 (0)40 7410 59955. E-mail address: [email protected] (E.M. Bauch).

(Schultz, 2007; Tobler, 2005) but also upcoming threat (Bromberg-Martin, Matsumoto, & Hikosaka, 2010; Ilango et al., 2014; Lammel et al., 2012). More specifically, in monkeys, SN/VTA activity increased linearly with the cue’s probability to predict an aversive air puff to the eye and an appetitive juice drop, respectively (Matsumoto & Hikosaka, 2009). While there are several findings in animals supporting the involvement of the dopaminergic midbrain in anticipating aversive events, there is only little evidence in humans (Bauch, Rausch, & Bunzeck, 2014; Fairhurst, Wiech, Dunckley, & Tracey, 2007). In an initial attempt to bridge this apparent gap between both species, we previously used fMRI in humans (Bauch et al., 2014). As a main finding, we could show that activity in the SN/VTA linearly increases as a function of shock probability during the anticipation of aversive events (electric shocks to the hand). However, the precise underlying temporal and oscillatory nature of shock anticipation remains unclear due to the sluggish properties of the BOLD signal. In contrast, electrophysiological techniques such as EEG and MEG offer the possibility to investigate neural activity at the level

http://dx.doi.org/10.1016/j.nlm.2015.06.010 1074-7427/Ó 2015 Published by Elsevier Inc.

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 2 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148

E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx

of milliseconds, which enables us to dissociate between multiple processes and underlying neural mechanisms involved in pain anticipation. Previous studies showed that pain anticipation can influence early sensory components of electrical brain activity (Babiloni et al., 2003; Dillmann, Miltner, & Weiss, 2000; Miyazaki et al., 1994; Weymar et al., 2013) and nociceptive events modulate oscillatory power in a range of frequency bands (May et al., 2012; Mouraux, Guérit, & Plaghki, 2003; Ohara, Crone, Weiss, & Lenz, 2004; Ploner et al., 2006a; Pomper et al., 2013; Raij, Forss, Stancák, & Hari, 2004). For instance, oscillatory power in the entire beta frequency range [low beta (13–20 Hz), high beta (20–30 Hz)] have repeatedly been shown to decrease (i.e. beta band suppression) in response to painful stimuli relative to non-painful or less painful events (Hauck, Lorenz, & Engel, 2007; Mancini, Longo, Canzoneri, Vallar, & Haggard, 2013; Ploner et al., 2006a; Pomper et al., 2013; Senkowski, Kautz, Hauck, Zimmermann, & Engel, 2011; Stancˇák, Polácˇek, Vrána, & Mlynárˇ, 2007). In contrast, beta power increases in response to tonic pain stimuli (Chang, Arendt-Nielsen, & Chen, 2002; Chang, Arendt-Nielsen, Graven-Nielsen, & Chen, 2003; Lalo et al., 2007) or long phasic stimulation (400 ms stimulation duration, Worthen, Hobson, Hall, Aziz, & Furlong, 2011). Moreover, modulations in the lower frequency range (theta: 4–8 Hz; alpha: 9–12 Hz) have been linked to nociceptive processing (Domnick, Hauck, Casey, Engel, & Lorenz, 2009; Iannetti, Hughes, Lee, & Mouraux, 2008; Mouraux et al., 2003). However, compared to the delivery of the nociceptive stimulus, it is largely unknown whether beta and theta power also signal the anticipation of an aversive and painful event. Initial evidence of beta power (14–30 Hz) increases during the anticipation phase of nociceptive stimuli has been reported in an MEG study (Worthen et al., 2011), where beta power modulations have been interpreted as a binding mechanism of pain-associated processes between different brain regions. In contrast to aversive anticipation, there is increasing evidence that the anticipation of appetitive stimuli, such as reward, modulates oscillatory power in the theta and beta frequency range (Bunzeck, Guitart-Masip, Dolan, & Düzel, 2011; Doñamayor, Marco-Pallarés, Heldmann, Schoenfeld, & Münte, 2011; van Wingerden, Vinck, Lankelma, & Pennartz, 2010). More specifically, in line with animal studies (Fiorillo, Tobler, & Schultz, 2003), the human brain quickly responds to cues that predict monetary rewards (i.e. 100 ms after cue onset). Moreover, while oscillatory power in the theta (5–8 Hz) band linearly decreases as a function of reward probability, high beta power (20–30 Hz) increases with reward probability (Bunzeck et al., 2011). Here, following the rationale of similar coding strategies between reward and aversive processing (Bauch et al., 2014; Bromberg-Martin et al., 2010), we tested the hypothesis that oscillatory power in the theta and beta frequency range also signals the anticipation of aversive events in a linear fashion. Behaviorally, aversive stimuli such as painful, electric shocks can have beneficial effects on long-term memory. More specifically, when aversive events are applied briefly after encoding, they can increase recognition memory possibly via enhanced arousal (Schwarze, Bingel, & Sommer, 2012; see also Dunsmoor, Martin, & LaBar, 2012; McCullough & Yonelinas, 2013). In a more recent study, we (Bauch et al., 2014) showed that even the anticipation of nociceptive events influences recognition memory. Recollection – recognition memory for contextual details of the studied episode (Tulving, 2002) – was modulated by shock probability following an inverted u-shape function. In other words, recollection was best for cues predicting an upcoming shock with a shock probability of 50%. This quadratic effect was mimicked by encoding-related activity in the posterior hippocampus. In contrast, familiarity – a general feeling of knowing the event in

absence of contextual details (Tulving, 2002) – was linearly scaled as a function of shock probability, which was paralleled by a linear increase in the anterior parahippocampal gyrus. Here, we used MEG to investigate the temporal and oscillatory dynamics of pain anticipation as a function of probability and its link to declarative memory formation. In a classical conditioning paradigm, participants initially learned to associate three different picture frames with three different probabilities (0.2, 0.5 or 0.8) to predict an electric shock. During an incidental encoding task, subjects categorized a series of indoor/outdoor scene images that were surrounded by these picture frames and were followed by an electric shock consistent with the frame’s probability. Approximately 15 min after encoding, recognition memory for the scene images was tested using a modified remember/know recognition task (Fig. 1). We predicted that the anticipation of aversive electric shocks is processed rapidly and signaled in beta (low beta: 13–20 Hz; high beta: 20–30 Hz) and/or theta (4–8 Hz) oscillatory power as a function of shock probability, similar to the effects in animal studies (Matsumoto & Hikosaka, 2009) and the reward literature (Bunzeck et al., 2011). Finally, we expected temporal and oscillatory modulations of brain activity associated with successful memory encoding as a function of shock probability (Bauch et al., 2014). A common index for successful memory formation is the so called ‘DM-effect’ (difference due to later memory, Paller, Kutas, & Mayes, 1987), that refers to the difference between encoding-related brain activity of later remembered and forgotten items. Previous M/EEG studies showed theta power increases (Duzel, Penny, & Burgess, 2010) and beta power decreases (Hanslmayr, Spitzer, & Bäuml, 2009; Hanslmayr, Staudigl, & Fellner, 2012) for remembered relatively to forgotten items (i.e. DM-effect). Based on these findings and our recent fMRI study (Bauch et al., 2014), showing increased hippocampal DM-activations for 0.5 shock probability, we expected increased theta power and decreased beta power for the DM-effect in the 0.5 shock probability condition relative to DM-effects associated with 0.2 and 0.8 shock probability.

149

2. Materials and methods

185

2.1. Participants

186

43 healthy young adults participated in the present study (27 female, age range 20–34 years, mean 25 years, age range: 20– 34 years). All participants were right-handed and had normal or corrected-to-normal vision, no history of neurological, psychiatric, or medical disorders or any current medical problems. Each participant gave written informed consent according to the approval of the local ethics committee (medical association Hamburg).

187

2.2. Task

194

The entire experiment took place while the subjects were placed in the MEG-scanner. Individual pain thresholds were calibrated before the actual experiment started. Here, participants rated the intensity of the electric shock on a visual analog scale (VAS) ranging from 0 (i.e. electric stimulation is not perceptible) to 10 (i.e. electric stimulation is intolerable). A shock intensity of seven was used as nociceptive stimulus throughout the experiment. All participants took part in three consecutive phases: a conditioning phase, an encoding task and a memory recognition task. In the conditioning phase, 20 green, 20 blue and 20 red colored, rectangular picture frames were randomly intermixed and presented in central vision for 1.5 s on a gray background (Fig. 1). Participants implicitly learned that the color of the cue (i.e. green, blue or red picture frame) predicted an aversive electric shock with

195

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

188 189 190 191 192 193

196 197 198 199 200 201 202 203 204 205 206 207 208

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx

3

Fig. 1. Experimental design. During the conditioning phase, colored picture frames (red, blue, green) were presented for 1500 ms, followed by a fixation cross for 3000 ms (±500 ms). 1 s after the offset of a cue an electric shock was applied according to the frame’s probability (0.2, 0.5 or 0.8). Subsequently, participants encoded images of scenes that were surrounded by these colored frames while making an indoor/outdoor judgment. Again electric shocks were applied according to the frame’s probability. Around 15 min after encoding, memory for the scenes was incidentally tested using a modified remember/know recognition procedure (see Section 2.2 for details). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233

a specific probability (0.2, 0.5 or 0.8). In each trial, an electric shock to the back of the right hand was applied according to the frame’s probability 1 s after the offset of a cue. An electric shock consisted of a train of three pulses of 2 ms each intermittent with a 50 ms interval. The contingencies between color and shock probability were randomly assigned for each subject. A colored cue was always followed by a fixation cross in central vision (3000 ± 500 ms). In total, a trial took approximately 5 s. Volunteers were ask to indicate the color of the frame as quickly and accurately as possible by pressing one of three buttons (the contingency between color and button was counterbalanced across participants). The conditioning phase lasted approximately 6 min. During encoding, trial structure and timing was identical to the conditioning phase but additionally photographs of indoor or outdoor scenes were presented within each of the colored picture frames (Fig. 1). Importantly, during this phase, an electric shock to the back of the right hand was also applied according to the frame’s probability 1 s after the offset of a cue. The task was to judge the indoor/outdoor status as quickly and as accurately as possible by pressing one of two buttons using the right index or middle finger, respectively. The response buttons for the indoor/outdoor judgment was counterbalanced across participants. In total 210 scene images (i.e. 35 indoor and 35 outdoor scenes for each of the three frame colors/shock probabilities) were presented during the encoding phase, which was split into three

blocks of 70 images. The images were gray-scaled and normalized to a mean value of 127 (SD = 75, RGB-space 0–255). The encoding task was approximately 18 min long and participants were initially familiarized with the task in a short practice session. Around 15 min after the encoding session, subjects’ memory for the scene images was tested in an incidental recognition test (i.e. participants encountered scene images during encoding without knowing that their memory would subsequently be tested). All 210 studied (i.e. old) scene images were randomly intermixed with 70 unstudied (i.e. new) scene images (35 indoor and 35 outdoor scenes) and presented in central vision on a gray background for 1.5 s. Before the beginning of the recognition task, the shock electrode was removed from the back of the volunteers’ hand. Volunteers were asked to differentiate between old and new images according to a modified remember/know procedure to dissociate between recognition based on recollection and familiarity (for review see Yonelinas, 2002). According to dual process models, recognition memory can be linked to recollection or familiarity: recollection-based recognition is associated with the retrieval of contextual details of the studied episode and familiarity refers to the recognition in the absence of contextual information (Yonelinas, 1995). If subjects confidently remembered a studied image and could recollect any specific details about it, they made a ‘remember’ response. If the item was familiar and participants recognized the picture without recollecting any details, they

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 4

E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx

270

pressed the ‘know’ button. If they confidently classified an image as unstudied (i.e. new), they gave a ‘new’ response. To minimize contamination by successful guesses, subjects had to press the ‘guess’ button whenever they were not confident about the old/new status of the image. The mapping of the responding fingers (i.e. index finger, middle finger, ring finger and little finger of the right hand) and corresponding decisions (i.e. remember, know, new, guess) was randomized across subjects. The recognition test consisted of four blocks of 70 scene images. After finishing the recognition test, participants were asked to rate the unpleasantness of the electric shock on a VAS ranging from 0 (not unpleasant) to 10 (highly unpleasant).

271

2.3. Behavioral data analyses

272

285

Recognition memory performance was analyzed according to the assumption that recollection and familiarity are two independent recognition processes (Yonelinas, 1995). The probability of recollection for each shock probability was calculated by subtracting the proportion of false alarms (i.e. incorrect remember judgments for new items, Fa R) from the proportion of hits (i.e. correct remember judgment for studied items, R) [R Fa R]. Familiarity was estimated as the probability of ‘know’ responses (K) corrected for the probability of making ‘know’ responses to new items (Fa K) and corrected for the fact that ‘know’ responses were given in the absence of recollection: familiarity = (K Fa K)/(1 R) (Yonelinas, 2002). Greenhouse-Geisser corrected p and df values are reported whenever a factor had more than two levels (Keselman & Rogan, 1980).

286

2.4. MEG acquisition

287

304

MEG was recorded in a magnetically shielded chamber using a 275-channel CTF MEG-system with SQUID-based axial gradiometers (VSM MedTech Ltd., Couquitlam, BC, Canada). Four faulty channels (MLF21, MRO42, MRO11 and MLC17) were excluded from data analysis (i.e. 271 MEG channels remained). Continuous digitization of the signals was performed at a sampling rate of 1200 Hz. Online, data were low-pass filtered at 240 Hz. Only data from the encoding phase were preprocessed and analyzed with SPM8 (Wellcome Trust Centre for Neuroimaging, University College London, UK) and MATLAB (The MathWorks, Inc., Natwick, MA, USA). The major benefit of using SPM8 for MEG-data is that voxel-based images are used for the statistical analyses. The second-level analyses are based on general linear models and random field theory (Kiebel & Friston, 2004; Worsley, Taylor, Tomaiuolo, & Lerch, 2004) to account for multiple comparison problems. This analysis approach has been applied in several previous publications (Apitz & Bunzeck, 2013; Bunzeck et al., 2011; Eckart & Bunzeck, 2013; Henson, Mouchlianitis, & Friston, 2009).

305

2.5. ERF analysis

306

Data were offline filtered between 0.25 and 20 Hz to remove low and high-frequency noise and epoched from 100 ms before to 1000 ms after cue onset. The data were then down-sampled to 150 Hz and artifacts were detected using a thresholding technique to remove trials with signals above 2500 fT. In the next step, artifact-corrected trials were averaged across conditions. The averaged data were converted into Neuroimaging Informatics Technology Initiative (NIfTI) format. Here, one 3D image of channel space  time was generated for each condition and subject. The sensor locations were projected onto a plane followed by a linear interpolation to a 64  64 pixel grid (pixel size = 2.12  2.69 mm) to provide a 2D channel space. The time dimension was 166 samples of 6.67 ms per epoch. These images were smoothed using a

259 260 261 262 263 264 265 266 267 268 269

273 274 275 276 277 278 279 280 281 282 283 284

288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303

307 308 309 310 311 312 313 314 315 316 317 318

Gaussian kernel (full-width half-maximum, FWHM) of FWHM = 5  5  15 mm to compensate for the spatial and/or temporal variance between subjects and to account better for the random field theory. Two separate second-level random-effects analyses were performed on the ERF-data (i.e. 3D-images). The first analysis focused on linear pain anticipation effects with respect to the different shock probabilities (i.e. one-way ANOVA with the factor shock probability: 0.2/0.5/0.8; T-contrast: 1 0 1; see Section 1: Bauch et al., 2014; Bunzeck et al., 2011). The second ERF analysis was performed to capture differences of neural activity associated with successful memory formation with respect to the three different shock probabilities. This analysis was restricted to remember trials due to too few trials for the know condition. ERF waveforms were created for each subject and sensor by subtracting the averaged epochs associated with later remembered images from those related to subsequently forgotten images (i.e. DM) with regard to each shock probability (i.e. DM 0.2, DM 0.5, DM 0.8). In other words, the analysis consisted of the within-subject factor ‘DM shock probability’ resulting in a 1  3 ANOVA. Here, we tested for a quadratic effect (Bauch et al., 2014). All statistical parametric maps were initially thresholded at p < 0.001 (uncorrected) and corrected using a priori time-windows of interest (see Section 3.2 and 1).

319

2.6. Time–frequency decomposition

343

TF data were epoched from 450 before to 1000 ms after cue onset, baseline corrected relative to 250 ms before stimulus onset; low-pass filtered at 50 Hz, down-sampled at 150 Hz and artifact thresholded at 2500fT. Spectral decomposition was applied to the preprocessed data at a trial-by-trial level using Morlet wavelets (factor 7) in the frequency range of 4–30 Hz. Time–frequency power was then averaged for each probability condition (i.e. 0.2, 0.5, 0.8). For DM-effects, time–frequency power was averaged across all trials for each of the different shock probability conditions and subsequent memory performance (i.e. remember 0.2, remember 0.5, remember 0.8, forgotten 0.2, forgotten 0.5, forgotten 0.8). The TF spectrogram was then rescaled by dividing the power of the trial (p) by the power of the baseline (p_b) and taking the log of this ratio [Log R: (log(p/p_b))]. Finally, the TF data were converted into nifti format for each of the three frequency ranges of interest (theta: 4–8 Hz; low beta: 13–20 Hz and high beta: 20–30 Hz), which generated a 3D image of channel space  time (Kilner & Friston, 2010). The process of creating a 2D channel space was identical to the one used for the ERF analysis. The time dimension consisted of 218 samples per epoch with a length of 6.67 ms. Then, these images were smoothed using a Gaussian kernel of FWHM = 6  6  15. To test for linear effects of pain anticipation with respect to shock probability (Bauch et al., 2014; Bunzeck et al., 2011) TF data (for each frequency range of interest) were analyzed using a random-effects 3  1 ANOVA (0.2/0.5/0.8). To investigate DM-effects, we used a random-effects 1  3 ANOVA for each frequency range of interest consisting of the within-subject factor DM (i.e. remember minus forgotten) shock probability (DM 0.2, DM 0.5, DM 0.8). Again, the analysis was restricted to remember responses trials due to the limited number of trials for the know condition. For the MEG data analysis of the anticipation phase irrespective of recognition performance, the mean number of trials for high, medium and low probability were 64 (range: 41–70), 64 (41–69) and 64 (39–70), respectively. For the DM-analysis, the mean number of trials was relatively low due to low recognition performance (high remember: 24 [14–31]; medium remember: 23 [14–31]; low remember: 23 [14–40]; high forgotten: 29 [19–43]; medium

344

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342

345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 5

E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx 383 384 385 386

forgotten: 32 [24–46]; low forgotten: 28 [15–40]). Only participants with at least 14 trials per condition (right site of the plot in Fig. 2) and a positive correlation effect were included in the DM-analysis (n = 14).

387

3. Results

388

3.1. Behavioral results

389

Discrimination performance between indoor and outdoor scenes for shock probabilities of 0.2, 0.5 and 0.8 during encoding was high [mean accuracy: 0.97 (SD = 0.03), 0.96 (SD = 0.03), 0.97 (SD = 0.04)]. A 3  1 ANOVA with shock probability as within-subject factor revealed no difference between conditions (p = 0.140). Similarly, there was no significant difference (p = 0.116) in reaction times between shock probabilities [0.2: 760 ms (SD = 106 ms), 0.5: 755 ms (SD = 100 ms); 0.8: 739 ms (SD = 110 ms)]. Recognition memory performance at retrieval is listed in Tables 1 and 2; all analyses are based on corrected hit rates (see Section 2.3). Although recognition performance was rather low, remember and know responses significantly differed from chance level (0.25) (T(42) = 10.35; p < 0.0001; T(42) = 7.04; p < 0.0001). Two 3  1 ANOVAs with the within-subject factors shock probability (0.2/0.5/0.8) on recollection and familiarity (see Section 2.3 and Bauch et al., 2014), respectively, did not reveal significant effects (p = 0.340 and p = 0.540). In a next step, we computed a correlation between the unpleasantness rating and the expected quadratic shock probability effect on the recollection rate (i.e. averaged value across recollection rate in 0.2 and 0.8 probability condition minus recollection rate in 0.5 probability condition) to understand this null finding. Note that previous studies already showed evidence that psychological factors, such as arousal and anxiety, can modulate subsequent memory performance (Anderson, Yamaguchi, Grabski, & Lacka, 2006; Cahill & McGaugh, 1998; Callan & Schweighofer, 2008; Murty, Labar, Hamilton, & Adcock, 2011), which motivated this post hoc analysis. Similarly, we calculated a correlation analysis between the predicted linear shock probability effect on the familiarity rate (i.e. average value across familiarity rate in 0.2 and 0.5 probability conditions minus familiarity rate in 0.8 probability condition) and the unpleasantness rating. The analyses revealed a quadratic relationship between recollection-based performance and the unpleasantness rating (r = 0.40; p = 0.008). That means, the more unpleasant the rating of the electric shock, the more quadratic was the probability effect on recollection, see Fig. 2. Note that the effect was still evident

390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425

Fig. 2. Positive correlation between the quadratic recollection effect (difference between recollection rate in 0.5 probability condition and the average recollection rate across 0.2 and 0.8 probability condition) and unpleasantness ratings of the electric shock (n = 43).

Table 1 Recognition performance for old (studied) indoor and outdoor scenes for all three shock probabilities (0.2, 0.5 and 0.8; n = 43). Probability

Remember

Know

Misses

Guesses

0.8 0.5 0.2

0.30 (0.14) 0.31 (0.15) 0.31 (0.15)

0.21 (0.12) 0.21 (0.13) 0.22 (0.12)

0.21 (0.14) 0.19 (0.12) 0.19 (0.10)

0.28 (0.15) 0.29 (0.16) 0.28 (0.14)

Values represent proportions of remember, know, miss and guess responses to old (studied) pictures across-subject means (SD).

Table 2 Performance for new indoor/outdoor scenes in recognition test (n = 43). CR

FA – remember

FA – know

Guess

0.52 (0.19)

0.03 (0.04)

0.10 (0.09)

0.35 (0.18)

Values represent proportions of correct rejections (CR), false alarms (FA) and guess responses to new (unstudied) scenes across-subject means (SD).

across those participants (n = 21) with enough trials (P14) that were later included in the DM-analysis of the MEG data (r = 0.50; p = 0.010). Finally, there was no linear (or quadratic) relationship between familiarity responses (Bauch et al., 2014; average of familiarity score across 0.1 and 0.5 minus the familiarity score in 0.8 condition) and the unpleasantness rating (linear: r = 0.180; p = 0.248; quadratic: r = 0.050; p = 0.749).

426

3.2. MEG results

433

Both ERF and TF analyses were initially explored at an uncorrected level of p = 0.001 followed by family-wise error (FWE) correction for multiple comparisons. The time window for multiple comparisons for the effects of pain anticipation effects was set to 100–300 ms after cue onset (Bunzeck et al., 2011; Senkowski et al., 2011). For the ERF DM-effects, we chose a time window from 400 ms to 500 ms after cue onset which was consistent with the maximum difference across conditions after visual inspection. DM-effects have been shown to vary in time with respect to their latency and onset (Paller et al., 1987; Voss & Paller, 2009). FWE corrections were applied to the time (z) but not space dimension (x, y). The TF analysis in the low beta frequency range (13–20 Hz) resulted in a significant linear pain anticipation effect at left temporal sensors (t-contrast 1 0 1; FWE-corrected p = 0.038; peak 103 ms; 316 voxels; see Fig. 3). TF analyses in the theta and high beta range as well as ERF analyses testing for linear pain anticipation effects did not reveal significant results (uncorrected threshold p < 0.001). In the next step, we computed DM-analyses for ERFs and TF-data. In accordance with the behavioral findings, DM-effects were calculated for those participants who showed a positive relationship between unpleasantness rating and the expected quadratic shock probability effect on recollection (see Section 3.1). We included only participants with more than 14 trials per condition (n = 14). Importantly, ERF analyses resulted in a significant quadratic effect (F-contrast: 1 2 1) at left frontal sensors (FWE-corrected p = 0.029; peak 447 ms; 2511 voxels; see Fig. 4). That means, the ERF difference for remembered vs. forgotten items showed a significant positive deflection for the 0.5 probability condition, while this difference was negative in the 0.2 and 0.8 probability conditions (Fig. 4). For a full depiction of the ERFs associated with remembered and forgotten items for all three probabilities see Fig. 5. There were no DM-effect differences between probability conditions in the TF domain. However, across all three probability conditions, there was a decrease for remembered scenes as compared

434

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

427 428 429 430 431 432

435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 6

E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx

Fig. 3. Results of time–frequency analyses during shock anticipation. Low beta power (13–20 Hz) increased linearly with shock probability at left temporal sensors (x = 40, y = 49; nearest channel: MLT37) with a peak at 103 ms. Left column depicts statistical parametric maps of the t-statistic. Middle graph shows low beta power for each probability condition at the peak time point that was extracted from voxel space (error bars denote one SEM.); right graph shows the time–frequency plot for the nearest channel (see Sections 2.4 and 2.6).

Fig. 4. Encoding-associated event-related magnetic fields. DM-related ERF analyses revealed a positive quadratic effect at left frontal sensors (x = 34, y = 16; nearest channel: MLT13). Left graph shows statistical parametric maps of the F-statistic. Second column shows the activity for each shock probability at the peak time point that was extracted from voxel space. Third column shows the time course of the effects for the three shock probabilities as extracted from voxel space (see Section 2.5).

Fig. 5. Encoding-associated event-related magnetic fields separately for the three shock probabilities (02, 0.5, 0.2). For the 0.2 and 0.8 probability condition, ERFs associated with remembered scenes showed a more negative-deflection as compared to the ERFs associated with forgotten scenes. In contrast, for the 0.5 shock probability condition ERFs associated with remembered scenes were more positive than for forgotten scenes.

470 471 472

to forgotten scenes for both theta (peak: 288 ms; voxels: 7119; pFWE = 0.031) and beta power (peak: 277 ms; voxels: 1519; pFWE = 0.046).

473

4. Discussion

474

We investigated the temporal and oscillatory dynamics of pain anticipation for different shock probabilities and its influence on

475

subsequent long-term memory. Behaviorally, across all participants, subsequent recognition memory was not modulated by shock probabilities during pain anticipation (i.e. encoding). However, when considering the degree of unpleasantness, shock probability modulated recollection-based recognition following an inverted u-shape function. More precisely, recollection was highest for scenes that were encoded in the context of 50% shock probability, and this effect was most pronounced when the electric

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

476 477 478 479 480 481 482 483

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549

shock was rated as unpleasant, which is partly compatible with our previous work (Bauch et al., 2014). This behavioral effect was mimicked in the ERFs starting at approximately 400 ms after cue onset lasting until around 600 ms. Here, the DM-effect associated with a medium shock probability showed a more positive deflection at left frontal sensors as compared to low and high shock probability conditions. Studies in animals (Horvitz, 2000; Ilango et al., 2014; Lammel et al., 2012; Ungless, 2004; Zweifel et al., 2011) and recent work in humans (Bauch et al., 2014; Guitart-Masip, Bunzeck, Stephan, Dolan, & Düzel, 2010) showed evidence that the processing of aversive and appetitive events rely on common neural structures. In particular, the dopaminergic mesolimbic system, including the SN/VTA, responds to cues that signal reward (Guitart-Masip et al., 2010) and electric shocks (Bauch et al., 2014). Our findings expand this previous work (Bauch et al., 2014; Matsumoto & Hikosaka, 2009) by showing that the anticipation of aversive events (i.e. electric shock) is signaled in oscillatory power following similar coding strategies as for monetary reward anticipation (Bunzeck et al., 2011). More precisely, we show that oscillatory power specifically in the low beta frequency range (13–20 Hz) linearly increased between 100 and 300 ms after cue onset at left temporal sensors as a function of shock probability (Fig. 3). A recent fMRI study, in which we used the same paradigm showed linear increases to pain anticipation in the dopaminergic midbrain (i.e. SN/VTA) (Bauch et al., 2014). Here, it is more likely that the beta effect originates in the frontal cortex or medial temporal lobe, which has direct connections to other brain regions of the dopaminergic mesolimbic system, such as the ventral striatum and SN/VTA (Lisman & Grace, 2005). Physiologically, oscillations in the lower frequency range (<30 Hz), such as beta, are thought to play a crucial role in integrating information across distant brain structures (Donner & Siegel, 2011). In the same vein, beta oscillations have been suggested to bind information from salient (Brenner et al., 2009) and rewarding (Bunzeck et al., 2011; HajiHosseini, Rodríguez-Fornells, & Marco-Pallarés, 2012) events originating in cortico-subcortical networks to guide learning and memory formation (Marco-Pallarés, Münte, & Rodríguez-Fornells, 2015). Thus, in the present study, low beta oscillations may reflect the binding of neural assemblies that signal motivational and aversive information between the prefrontal cortex and other dopaminergic mesolimbic structures (e.g. SN/VTA). One possibility to more directly investigate this hypothesis is combined EEG/fMRI which allows to integrate the high spatial and temporal resolution of both methods (Huster, Debener, Eichele, & Herrmann, 2012). Importantly, we would like to point out that our findings of a linear increase in low beta power (13–20 Hz) to aversive events are only partly compatible with a previous MEG study showing increases in high beta power (20–30 Hz) at fronto-central sensors with increasing reward probability (Bunzeck et al., 2011). Therefore, processing aversive and appetitive events may depend on similar but not identical neural mechanisms (Bromberg-Martin et al., 2010). Future research needs to clarify this open question on the basis of a direct comparison between appetitive and aversive information within the same experiment. Regarding long-term memory, we can extend our previous work (Bauch et al., 2014) by showing that there is no canonical effect of shock anticipation on learning. Instead, subjective experience of pleasantness seems to be a crucial psychological variable. More precisely, although the subjective intensity of the shocks was identical between subjects (7 on a 1–10 VAS) at the beginning of the experiment, there were different reports of subjective pleasantness of the electric shock at the end of the experiment. Interestingly, the predicted inverted u-shape relationship between shock probability and recollection was most pronounced in subjects with high unpleasantness rating (Fig. 2). Thus, high

7

unpleasantness of the electric shock, which most likely was associated with higher arousal throughout the entire encoding phase, interacted with memory-related effects of pain anticipation. Indeed, previous studies show evidence that high arousal due to aversive stimuli was paralleled by an elevation in corticosteroid hormones, which, in turn, can increase recognition performance, particularly hippocampus-dependent recollection rate (Anderson et al., 2006; Cahill, Gorski, & Le, 2003; Cahill & McGaugh, 1998; McGaugh, 2004; McGaugh, Cahill, & Roozendaal, 1996; McGaugh & Roozendaal, 2002). Since the quadratic effect of shock probability on recollection has been associated with hippocampal activity in a previous fMRI study (Bauch et al., 2014), it may be possible that increased arousal levels may have boosted the subtle probability effect on recollection observed here (i.e. memory benefit for 0.5 shock probability). Taken together, the subtle effect of shock probability on subsequent recollection seems to depend on psychological factors that relate to unpleasantness and possibly elevated arousal. Surprisingly and not in line with our previous study (Bauch et al., 2014), there was no modulation of familiarity-based recognition by shock probability or unpleasantness of the electric shock. This is most likely due to low statistical power as indexed by significantly lower familiarity rates as compared to recollection rate (t(42) = 2.89; p = 0.006) across all participants. Alternatively, if unpleasantness of the electric shocks led to increased corticosteroid hormones and arousal, which in turn may particularly influence hippocampus-related recollection, it may not be surprising that familiarity (associated with perirhinal and entorhinal cortex) is not affected by unpleasantness of the electric shock. We suggest that future studies apply paradigms that better tap into familiarity-based recognition in order to investigate the relationship between pain anticipation and familiarity-based recognition more thoroughly. At the neural level, the quadratic effect for recollection-based memory was paralleled by left frontal encoding-related ERFs (DM-effect) around 450 ms after stimulus onset. This is consistent with previous EEG studies showing DM-effects to be maximal at fronto-central scalp sites in a similar time period (Duarte, Ranganath, Winward, Hayward, & Knight, 2004; Otten & Rugg, 2001) and functional MRI studies that repeatedly found an involvement of the PFC in successful memory formation (Blumenfeld & Ranganath, 2007). Moreover, fronto-central subsequent memory effects between 400 and 600 ms have been associated with emotional (negative, positive) stimuli (Dolcos & Cabeza, 2002). Therefore, the frontal distribution of our DM-effect for 0.5 shock probability may be linked with prefrontal generators that have connections to other mesolimbic regions, possibly the hippocampus (Bauch et al., 2014; Lisman & Grace, 2005), to benefit successful memory encoding. Although this interpretation resonates well with previous reports, our findings need to be handled with caution given the relatively low number of subjects (n = 14). Interestingly, the DM-effect was reversed for the 0.2 and 0.8 shock probability as compared to the 0.5 condition (Fig. 5). Differences in polarity between ERPs/ERFs indicate the engagement of qualitatively different cognitive processes and different underlying neuronal populations (Otten & Rugg, 2004). For instance, previous EEG studies reported both positive- and negative-going DM-effects that were associated with differences in encoding tasks (focus on semantic or shallow information), encoded material or retrieval requirements (e.g. Bridger & Wilding, 2010). Therefore, our findings suggest that qualitatively different processes and neuronal populations may be engaged between encoded scenes for 0.5 and the other two shock probabilities. Finally, we did not fully replicate our findings from a recent behavioral study (Bauch et al., 2014), which may speak for a less

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 8

E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx

632

robust effect of shock probability on learning. In the MEG scanner, shock probability did not modulate recognition memory (recollection and familiarity) per se, but was influenced by the unpleasantness of the shock. Similarly to a previous fMRI study (Bauch et al., 2014), the MEG scanner may induce increased arousal and/or anxiety, which increases adrenal stress which, in turn, may interfere with the effects of shock anticipation on recognition memory (Cooke, Peel, Shaw, & Senior, 2007; Eatough et al., 2009; Muehlhan et al., 2011; Peters et al., 2011; Tessner et al., 2006). The impact of shock unpleasantness on recollection in the present study is in line with this interpretation. Alternatively, in the current study recognition memory was rather low and the familiarity analyses were based on a low number of trials. Therefore, we suggest further studies with higher statistical power to validate our findings. Here, the key to success might relate to a refinement of the experimental design, for instance, by using more memorable stimulus material.

633

5. Conclusions

634

646

Taken together, we show that the anticipation of aversive events can improve recollection-based recognition memory following an inverted u-shape function. Importantly, this effect seems to be modulated by the subjective experience of pleasantness and, in the case of higher unpleasantness, quadratic memory modulations were mimicked by ERFs at left frontal sensors at around 450 ms after stimulus onset. Moreover, oscillatory low beta (13– 20 Hz) power was linearly scaled at left temporal sensors as a function of shock probability, indicating that low beta power may integrate neural assemblies representing prospective, aversive information. At a more general level, our findings indicate common, but also distinct neural coding parameters of aversive and appetitive processing.

647

6. Uncited reference

616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631

635 636 637 638 639 640 641 642 643 644 645

648

Ploner, Gross, Timmermann, Pollok, and Schnitzler (2006c).

649

Acknowledgments

650 652

This work was supported by grants from the German Research Foundation (Deutsche Forschungsgemeinschaft, BU 2670/1-1) to NB and Hamburg state cluster of excellence (neurodapt!).

653

References

654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678

Anderson, A. K., Yamaguchi, Y., Grabski, W., & Lacka, D. (2006). Emotional memories are not all created equal: Evidence for selective memory enhancement. Learning & Memory, 13(6). Apitz, T., & Bunzeck, N. (2013). Dopamine controls the neural dynamics of memory signals and retrieval accuracy. Neuropsychopharmacology, 38(12), 2409–2417. Babiloni, C., Brancucci, A., Babiloni, F., Capotosto, P., Carducci, F., Cincotti, F., et al. (2003). Anticipatory cortical responses during the expectancy of a predictable painful stimulation. A high-resolution electroencephalography study. European Journal of Neuroscience, 18(6). Bauch, E. M., Rausch, V. H., & Bunzeck, N. (2014). Pain anticipation recruits the mesolimbic system and differentially modulates subsequent recognition memory. Human Brain Mapping, 35(9), 4594–4606. Blumenfeld, R. S., & Ranganath, C. (2007). Prefrontal cortex and long-term memory encoding: An integrative review of findings from neuropsychology and neuroimaging. The Neuroscientist, 13(3), 280–291. Brenner, C. A., Kieffaber, P. D., Clementz, B. A., Johannesen, J. K., Shekhar, A., O’Donnell, B. F., et al. (2009). Event-related potential abnormalities in schizophrenia: A failure to ‘‘gate in’’ salient information? Schizophrenia Research, 113(2–3), 332–338. Bridger, E. K., & Wilding, E. L. (2010). Requirements at retrieval modulate subsequent memory effects: An event-related potential study. Cognitive Neuroscience, 1(4), 254–260. Bromberg-Martin, E. S., Matsumoto, M., & Hikosaka, O. (2010). Dopamine in motivational control: Rewarding, aversive, and alerting. Neuron, 68(5), 815–834.

651

Bunzeck, N., Guitart-Masip, M., Dolan, R. J., & Düzel, E. (2011). Contextual novelty modulates the neural dynamics of reward anticipation. The Journal of Neuroscience, 31(36), 12816–12822. Cahill, L., Gorski, L., & Le, K. (2003). Enhanced human memory consolidation with post-learning stress: Interaction with the degree of arousal at encoding. Learning & Memory, 10(4), 270–274. Cahill, L., & McGaugh, J. L. (1998). Mechanisms of emotional arousal and lasting declarative memory. Trends in Neurosciences, 21(7), 294–299. http://dx.doi.org/ 10.1016/S0166-2236(97)01214-9. Callan, D. E., & Schweighofer, N. (2008). Positive and negative modulation of word learning by reward anticipation. Human Brain Mapping, 29(2), 237–249. Chang, P. F., Arendt-Nielsen, L., & Chen, A. C. N. (2002). Dynamic changes and spatial correlation of EEG activities during cold pressor test in man. Brain Research Bulletin, 57(5), 667–675. Chang, P.-F., Arendt-Nielsen, L., Graven-Nielsen, T., & Chen, A. C. N. (2003). Psychophysical and EEG responses to repeated experimental muscle pain in humans: Pain intensity encodes EEG activity. Brain Research Bulletin, 59(6), 533–543. Cooke, R., Peel, E., Shaw, R. L., & Senior, C. (2007). The neuroimaging research process from the participants’ perspective. International Journal of Psychophysiology, 63(2), 152–158. Dillmann, J., Miltner, W. H. R., & Weiss, T. (2000). The influence of semantic priming on event-related potentials to painful laser-heat stimuli in humans. Neuroscience Letters, 284(1–2), 53–56. Dolan, R. J., Heinze, H. J., Hurlemann, R., & Hinrichs, H. (2006). Magnetoencephalography (MEG) determined temporal modulation of visual and auditory sensory processing in the context of classical conditioning to faces. NeuroImage, 32(2), 778–789. Dolcos, F., & Cabeza, R. (2002). Event-related potentials of emotional memory: Encoding pleasant, unpleasant, and neutral pictures. Cognitive, Affective, & Behavioral Neuroscience, 2(3), 252–263. Domnick, C., Hauck, M., Casey, K. L., Engel, A. K., & Lorenz, J. (2009). C-fiber-related EEG-oscillations induced by laser radiant heat stimulation of capsaicin-treated skin. Journal of Pain Research, 2, 49–56. Doñamayor, N., Marco-Pallarés, J., Heldmann, M., Schoenfeld, M. A., & Münte, T. F. (2011). Temporal dynamics of reward processing revealed by magnetoencephalography. Human Brain Mapping, 32(12), 2228–2240. Donner, T. H., & Siegel, M. (2011). A framework for local cortical oscillation patterns. Trends in Cognitive Sciences, 15(5), 191–199. Duarte, A., Ranganath, C., Winward, L., Hayward, D., & Knight, R. T. (2004). Dissociable neural correlates for familiarity and recollection during the encoding and retrieval of pictures. Cognitive Brain Research, 18(3), 255–272. Dunsmoor, J. E., Martin, A., & LaBar, K. S. (2012). Role of conceptual knowledge in learning and retention of conditioned fear. Biological Psychology, 89(2), 300–305. Duzel, E., Penny, W. D., & Burgess, N. (2010). Brain oscillations and memory. Current Opinion in Neurobiology, 20(2), 143–149. Eckart, C., & Bunzeck, N. (2013). Dopamine modulates processing speed in the human mesolimbic system. NeuroImage, 66, 293–300. Fairhurst, M., Wiech, K., Dunckley, P., & Tracey, I. (2007). Anticipatory brainstem activity predicts neural processing of pain in humans. Pain, 128(1–2), 101–110. Fiorillo, C. D., Tobler, P. N., & Schultz, W. (2003). Discrete coding of reward probability and uncertainty by dopamine neurons. Science, 299(5614), 1898–1902. Garcia-Larrea, L., Frot, M., & Valeriani, M. (2003). Brain generators of laser-evoked potentials: From dipoles to functional significance. Neurophysiologie Clinique/ Clinical Neurophysiology, 33(6), 279–292. Guitart-Masip, M., Bunzeck, N., Stephan, K. E., Dolan, R. J., & Düzel, E. (2010). Contextual novelty changes reward representations in the striatum. The Journal of Neuroscience, 30(5), 1721–1726. HajiHosseini, A., Rodríguez-Fornells, A., & Marco-Pallarés, J. (2012). The role of betagamma oscillations in unexpected rewards processing. NeuroImage, 60(3), 1678–1685. Hanslmayr, S., Spitzer, B., & Bäuml, K.-H. (2009). Brain oscillations dissociate between semantic and nonsemantic encoding of episodic memories. Cerebral Cortex, 19(7), 1631–1640. Hanslmayr, S., Staudigl, T., & Fellner, M.-C. (2012). Oscillatory power decreases and long-term memory: The information via desynchronization hypothesis. Frontiers in Human Neuroscience, 6, 74. Hauck, M., Lorenz, J., & Engel, A. K. (2007). Attention to painful stimulation enhances c-band activity and synchronization in human sensorimotor cortex. The Journal of Neuroscience, 27(35), 9270–9277. Henson, R. N., Mouchlianitis, E., & Friston, K. J. (2009). MEG and EEG data fusion: Simultaneous localisation of face-evoked responses. NeuroImage, 47(2), 581–589. Horvitz, J. (2000). Mesolimbocortical and nigrostriatal dopamine responses to salient non-reward events. Neuroscience, 96(4), 651–656. Huster, R. J., Debener, S., Eichele, T., & Herrmann, C. S. (2012). Methods for simultaneous EEG-fMRI: An Introductory review. The Journal of Neuroscience, 32(18), 6053–6060. Iannetti, G. D., Hughes, N. P., Lee, M. C., & Mouraux, A. (2008). Determinants of laserevoked EEG responses: Pain perception or stimulus saliency? Journal of Neurophysiology, 100(2), 815–828. Iannetti, G. D., Zambreanu, L., Cruccu, G., & Tracey, I. (2005). Operculoinsular cortex encodes pain intensity at the earliest stages of cortical processing as indicated by amplitude of laser-evoked potentials in humans. Neuroscience, 131(1), 199–208.

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010

679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747

748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764

YNLME 6276

No. of Pages 9, Model 5G

26 June 2015 E.M. Bauch, N. Bunzeck / Neurobiology of Learning and Memory xxx (2015) xxx–xxx 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820

Ilango, A., Kesner, A. J., Keller, K. L., Stuber, G. D., Bonci, A., & Ikemoto, S. (2014). Similar roles of substantia nigra and ventral tegmental dopamine neurons in reward and aversion. The Journal of Neuroscience, 34(3), 817–822. Keselman, H. J., & Rogan, J. C. (1980). Repeated measures F tests and psychophysiological research: Controlling the number of false positives. Psychophysiology, 17(5), 499–503. Kiebel, S. J., & Friston, K. J. (2004). Statistical parametric mapping for event-related potentials (II): A hierarchical temporal model. NeuroImage, 22(2), 503–520. Kilner, J. M., & Friston, K. J. (2010). Topological inference for EEG and MEG. The Annals of Applied Statistics, 4(3), 1272–1290. Lalo, E., Gilbertson, T., Doyle, L., Lazzaro, V. D., Cioni, B., & Brown, P. (2007). Phasic increases in cortical beta activity are associated with alterations in sensory processing in the human. Experimental Brain Research, 177(1), 137–145. Lammel, S., Lim, B. K., Ran, C., Huang, K. W., Betley, M. J., Tye, K. M., et al. (2012). Input-specific control of reward and aversion in the ventral tegmental area. Nature, 491(7423), 212–217. Lisman, J. E., & Grace, A. A. (2005). The hippocampal-VTA loop: Controlling the entry of information into long-term memory. Neuron, 46(5), 703–713. Mancini, F., Longo, M. R., Canzoneri, E., Vallar, G., & Haggard, P. (2013). Changes in cortical oscillations linked to multisensory modulation of nociception. European Journal of Neuroscience, 37(5), 768–776. Marco-Pallarés, J., Münte, T. F., & Rodríguez-Fornells, A. (2015). The role of highfrequency oscillatory activity in reward processing and learning. Neuroscience & Biobehavioral Reviews, 49, 1–7. Matsumoto, M., & Hikosaka, O. (2009). Two types of dopamine neuron distinctly convey positive and negative motivational signals. Nature, 459(7248), 837–841. May, E. S., Butz, M., Kahlbrock, N., Hoogenboom, N., Brenner, M., & Schnitzler, A. (2012). Pre- and post-stimulus alpha activity shows differential modulation with spatial attention during the processing of pain. NeuroImage, 62(3), 1965–1974. McCullough, A. M., & Yonelinas, A. P. (2013). Cold-pressor stress after learning enhances familiarity-based recognition memory in men. Neurobiology of Learning and Memory, 106, 11–17. McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annual Review of Neuroscience, 27(1), 1–28. McGaugh, J. L., Cahill, L., & Roozendaal, B. (1996). Involvement of the amygdala in memory storage: Interaction with other brain systems. Proceedings of the National Academy of Sciences, 93(24), 13508–13514. McGaugh, J. L., & Roozendaal, B. (2002). Role of adrenal stress hormones in forming lasting memories in the brain. Current Opinion in Neurobiology, 12(2), 205–210. Miyazaki, M., Shibasaki, H., Kanda, M., Xu, X., Shindo, K., Honda, M., et al. (1994). Generator mechanism of pain-related evoked potentials following CO2 laser stimulation of the hand: Scalp topography and effect of predictive warning signal. Journal of Clinical Neurophysiology, 11(2). Mouraux, A., Guérit, J. M., & Plaghki, L. (2003). Non-phase locked electroencephalogram (EEG) responses to CO2 laser skin stimulations may reflect central interactions between Ao- and C-fibre afferent volleys. Clinical Neurophysiology, 114(4), 710–722. Murty, V. P., Labar, K. S., Hamilton, D. A., & Adcock, R. A. (2011). Is all motivation good for learning? Dissociable influences of approach and avoidance motivation in declarative memory. Learning & Memory, 18(11), 712–717. Ohara, S., Crone, N. E., Weiss, N., & Lenz, F. A. (2004). Attention to a painful cutaneous laser stimulus modulates electrocorticographic event-related desynchronization in humans. Clinical Neurophysiology, 115(7), 1641–1652. Otten, L. J., & Rugg, M. D. (2001). Electrophysiological correlates of memory encoding are task-dependent. Cognitive Brain Research, 12(1), 11–18.

9

Paller, K. A., Kutas, M., & Mayes, A. R. (1987). Neural correlates of encoding in an incidental learning paradigm. Electroencephalography and Clinical Neurophysiology, 67(4), 360–371. Pizzagalli, D. A., Greischar, L. L., & Davidson, R. J. (2003). Spatio-temporal dynamics of brain mechanisms in aversive classical conditioning: High-density eventrelated potential and brain electrical tomography analyses. Neuropsychologia, 41(2), 184–194. Ploner, M., Gross, J., Timmermann, L., Pollok, B., & Schnitzler, A. (2006a). Oscillatory activity reflects the excitability of the human somatosensory system. NeuroImage, 32(3), 1231–1236. Ploner, M., Gross, J., Timmermann, L., Pollok, B., & Schnitzler, A. (2006c). Pain suppresses spontaneous brain rhythms. Cerebral Cortex, 16(4), 537–540. Pomper, U., Höfle, M., Hauck, M., Kathmann, N., Engel, A. K., & Senkowski, D. (2013). Crossmodal bias of visual input on pain perception and pain-induced beta activity. NeuroImage, 66, 469–478. Raij, T. T., Forss, N., Stancák, A., & Hari, R. (2004). Modulation of motor-cortex oscillatory activity by painful Ad- and C-fiber stimuli. NeuroImage, 23(2), 569–573. Schultz, W. (2007). Behavioral dopamine signals. Trends in Neurosciences, 30(5), 203–210. Schwarze, U., Bingel, U., & Sommer, T. (2012). Event-related nociceptive arousal enhances memory consolidation for neutral scenes. The Journal of Neuroscience, 32(4), 1481–1487. Senkowski, D., Kautz, J., Hauck, M., Zimmermann, R., & Engel, A. K. (2011). Emotional facial expressions modulate pain-induced beta and gamma oscillations in sensorimotor cortex. The Journal of Neuroscience, 31(41), 14542–14550. Stancˇák, A., Polácˇek, H., Vrána, J., & Mlynárˇ, J. (2007). Cortical oscillatory changes during warming and heating in humans. Neuroscience, 147(3), 842–852. Tobler, P. N. (2005). Adaptive coding of reward value by dopamine neurons. Science, 307(5715), 1642–1645. Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53(1), 1–25. Ungless, M. A. (2004). Dopamine: The salient issue. Trends in Neurosciences, 27(12), 702–706. van Wingerden, M., Vinck, M., Lankelma, J., & Pennartz, C. M. A. (2010). Theta-band phase locking of orbitofrontal neurons during reward expectancy. The Journal of Neuroscience, 30(20), 7078–7087. Voss, J. L., & Paller, K. A. (2009). Remembering and knowing: Electrophysiological distinctions at encoding but not retrieval. NeuroImage, 46(1), 280–289. Weymar, M., Bradley, M. M., Hamm, A. O., & Lang, P. J. (2013). When fear forms memories: Threat of shock and brain potentials during encoding and recognition. Cortex, 49(3), 819–826. Worsley, K. J., Taylor, J. E., Tomaiuolo, F., & Lerch, J. (2004). Unified univariate and multivariate random field theory. NeuroImage, 23(Suppl. 1), S189–S195. Worthen, S. F., Hobson, A. R., Hall, S. D., Aziz, Q., & Furlong, P. L. (2011). Primary and secondary somatosensory cortex responses to anticipation and pain: A magnetoencephalography study. European Journal of Neuroscience, 33(5), 946–959. Yonelinas, A. (1995). The relation between remembering and knowing as bases for recognition: Effects of size congruency. Journal of Memory and Language, 34(5), 622–643. Yonelinas, A. P. (2002). The nature of recollection and familiarity: A review of 30 years of research. Journal of Memory and Language, 46(3), 441–517. Zweifel, L. S., Fadok, J. P., Argilli, E., Garelick, M. G., Jones, G. L., Dickerson, T. M. K., et al. (2011). Activation of dopamine neurons is critical for aversive conditioning and prevention of generalized anxiety. Nature Neuroscience, 14(5), 620–626.

821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879

Please cite this article in press as: Bauch, E. M., & Bunzeck, N. Anticipation of electric shocks modulates low beta power and event-related fields during memory encoding. Neurobiology of Learning and Memory (2015), http://dx.doi.org/10.1016/j.nlm.2015.06.010