Deliberative Behaviors and Prefrontal-Hippocampal Coupling are Disrupted in a Rat Model of Fetal Alcohol Spectrum Disorders

29 Fetal alcohol spectrum disorders (FASDs) are characterized by a range of physical, cognitive, 30 and behavioral impairments. Determining how temporally specific alcohol exposure (AE) affects neural 31 circuits is crucial to understanding the FASD phenotype. Third trimester AE can be modeled in rats by 32 administering alcohol during the first two postnatal weeks, which damages the medial prefrontal cortex 33 (mPFC), thalamic nucleus reuniens, and hippocampus (HPC), structures whose functional interactions 34 are required for working memory and executive function. Therefore, we hypothesized that AE during this 35 period would impair working memory, disrupt choice behaviors, and alter mPFC-HPC oscillatory 36 synchrony. To test this hypothesis, we recorded local field potentials from the mPFC and dorsal HPC as 37 AE and sham intubated (SI) rats performed a spatial working memory task in adulthood and implemented 38 algorithms to detect vicarious trial and errors (VTEs), behaviors associated with deliberative decision-39 making. We found that, compared to the SI group, the AE group performed fewer VTEs and 40 demonstrated a disturbed relationship between VTEs and choice outcomes, while spatial working 41 memory was unimpaired. This behavioral disruption was accompanied by alterations to mPFC and HPC 42 oscillatory activity in the theta and beta bands, respectively, and a reduced prevalence of mPFC-HPC 43 synchronous events. When trained on multiple behavioral variables, a machine learning algorithm could 44 accurately predict whether rats were in the AE or SI group, thus characterizing a potential phenotype 45 following third trimester AE. Together, these findings indicate that third trimester AE disrupts mPFC-HPC 46 oscillatory interactions and choice behaviors. 47 48 Significance statement 49 Fetal alcohol spectrum disorders (FASDs) occur at an alarmingly high rate worldwide. Prenatal 50 alcohol exposure leads to significant perturbations in brain circuitry that are accompanied by cognitive 51 deficits, including disrupted executive functioning and working memory. These deficits stem from 52 structural changes within several key brain regions including the prefrontal cortex, thalamic nucleus 53 reuniens, and hippocampus. To better understand the cognitive deficits observed in FASD patients, we 54 employed a rodent model of alcohol exposure during the third trimester, a period when these regions are 55 especially vulnerable to alcohol-induced damage. We show that alcohol exposure disrupts choice 56 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint behaviors and prefrontal-hippocampal functional connectivity during a working memory task, identifying 57 the prefrontal-hippocampal network as a potential therapeutic target in FASD treatment. 58 59

60 Fetal alcohol spectrum disorders (FASDs) are the most common preventable cause of 61 developmental disability globally and are characterized by a range of physical defects and cognitive and 62 behavioral impairments, the extent of which are dependent on the timing of exposure to alcohol (AE) 63 (Coles, 1994; Hoyme et al., 2016; Mattson et al., 2019; Popova et al., 2023; Rasmussen, 2006). AE 64 during the brain growth spurt, which occurs during the third trimester in humans and the first two postnatal 65 weeks in rats (Dobbing & Sands, 1979), results in executive functioning deficits (Gursky et al., 2021; 66 Thomas et al., 1996), which are a hallmark of FASD (Mattson et al., 2019; Rasmussen, 2006). 67 The medial prefrontal cortex (mPFC), hippocampus (HPC), and their interaction are important for 68 memory-guided decision-making and are damaged after AE during the brain growth spurt (Bonthius & 69 West, 1991; Churchwell & Kesner, 2011; Floresco et al., 1997; Ikonomidou et al., 2000; Hamilton et al., 70 2010, 2017; Livy et al., 2003; Lawrence et al., 2012; Maharjan et al., 2018; Murawski et al., 2012; Otero 71 et al., 2012; Tran & Kelly, 2003; G.-W. Wang & Cai, 2006; Whitcher & Klintsova, 2008). The thalamic 72 nucleus reuniens mediates mPFC-HPC interactions during spatial working memory (Hallock et al., 2016) 73 and is also damaged after AE (Gursky et al., 2019, 2020), leading us to predict AE during this period 74 would impair spatial working memory. 75 The HPC, mPFC, and nucleus reuniens are implicated in choice behaviors known as vicarious 76 trial and errors (VTEs), which are thought to reflect deliberation and occur when rats pause and alternate 77 head movements towards choice options during decision-making (Bett et al., 2012; Blumenthal et al., 78 2011; Griesbach et al., 1998; Hu & Amsel, 1995; Papale et al., 2012; Kidder et al., 2021; Redish, 2016; 79 Schmidt et al., 2019; Stout et al., 2022; Tolman, 1939). VTEs emerge when flexible decision-making 80 strategies are favored, such as when task rules are switched, and diminish with increasing task 81 proficiency (Amemiya & Redish, 2016; Blumenthal et al., 2011; Griesbach et al., 1998; Hu & Amsel, 1995; 82 Papale et al., 2012; Redish, 2016; Steiner & Redish, 2012). HPC lesions or disruption (Bett et al., 2012; 83 Blumenthal et al., 2011; Griesbach et al., 1998; Hu & Amsel, 1995) and mPFC disruption (Kidder et al., 84 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint 2021; Schmidt et al., 2019) result in VTE reductions. Furthermore, nucleus reuniens inactivation 85 increases VTEs during consecutive choice error sequences, suggesting its importance for successful 86 deliberation (Stout et al., 2022). Consequently, we predicted VTE behaviors would be disrupted after AE. 87 As rats approach choice points, HPC ensembles alternate between representations of potential 88 choice trajectories ahead of the rat (Johnson & Redish, 2007, Kay et al., 2020; Tang et al., 2021). The 89 mPFC is hypothesized to evaluate these trajectories (Redish, 2016; J. X. Wang et al., 2015), which aligns 90 with PFC involvement in goal-directed and flexible behaviors (Miller & Cohen, 2001) and the increase in 91 mPFC-HPC oscillatory synchrony via theta rhythms (6-10 Hz oscillations in the local field potential; LFP) 92 during decision-making (Benchenane et al., 2010; Hallock et al., 2016; Jones & Wilson, 2005; O’Neill et 93 al., 2013). The nucleus reuniens has been shown to transfer trajectory-relevant information from mPFC to 94 HPC (Ito et al., 2015) and its inactivation reduces mPFC-HPC theta coherence (Hallock et al., 2016; Stout 95 et al., 2022), suggesting a critical role in mPFC-HPC interactions. Therefore, we predicted that AE would 96 lead to altered mPFC-HPC oscillatory activity during deliberation. 97 Our results show that AE during the brain growth spurt led to fewer VTEs in adulthood and 98 resulted in a dissociation between VTEs and subsequent task performance. Despite these disruptions, 99 task choice accuracy was unimpaired. We also demonstrate that mPFC-HPC physiology and functional 100 connectivity were disrupted in the AE group. Lastly, we show that a machine learning algorithm could 101 predict whether rats belonged to the AE or sham intubated (SI) group based on select behavioral 102 measures, therefore modeling a phenotype for third trimester AE. 103 104

105 Animal subjects 106 Subjects were Long Evans hooded rats (5 AE female, 6 AE male; 2 SI female, 5 SI male). Choice 107 accuracy over sessions analysis included an additional cohort of rats (9 AE female, 8 AE male; 9 SI 108 female, 13 SI male). Pregnant dams were obtained from Charles River (Wilmington, MA). Subjects were 109 generated from 10 litters and were born at the University of Delaware. The animal colony room was 110 temperature and humidity controlled and followed a light/dark cycle from 7 a.m.- 7 p.m. Rats had ad 111 libitum access to food and water until pretraining, when they were placed on mild food restriction to 112 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint maintain 90% of their original body weight. All animal procedures followed the University of Delaware 113 Institutional Animal Care and Use Committee (Animal Use Protocol #1177) and the NIH Guide for the 114 Care and Use of Laboratory Animals. See Figure 1A for the experimental timeline. 115 116 Animal generation and postnatal treatment 117 Pups were paw marked on postnatal day 3 with an injection of India black ink and were randomly 118 assigned to the AE or SI group. On postnatal days 4-9, pups in the AE group were administered 5.25 119 g/kg/day ethanol in a milk formula via intragastric intubation (divided between 2 doses at 9 a.m. and 11 120 a.m.). This procedure has been shown to result in a peak Blood Alcohol Concentration (BAC) of about 121 350 mg/dL (high dose) (Gursky et al., 2019, 2020, 2021) when measured 2 hours after the second 122 alcohol intubation. SI pups were intubated without any liquid to control for the stress effects of intubation. 123 To prevent weight loss, AE pups received a supplemental dose of milk formula 2 hours after the second 124 intubation on postnatal days 4-9 and an additional dose 4 hours after the second intubation on postnatal 125 day 4. Rats were ear punched for identification on postnatal day 9. All rats were housed with their dams 126 until postnatal day 23, when they were weaned and pair housed until surgery. 127 128 Behavior apparatus and testing room 129 Tasks were performed in a wooden T maze, which consisted of a central arm (116 cm x 10 cm), 130 two goal arms (56.5 cm x 10 cm), and two return arms (112 cm x 10 cm) with 6 cm high wooden walls. 131 Small weighing boats were attached at the end of each goal arm for food reward delivery. The start box at 132 the base of the maze consisted of a barstool with a dish attached on top. Visual cues were attached to a 133 black curtain that surrounded the room, which was dimly lit by 2 compact fluorescent bulbs. 134 135 Handling 136 After postnatal day 90, experimenters handled rats for 10 minutes/day for 5 days. After each 137 session, chocolate sprinkles were placed in the home cage to familiarize rats with the food reward of the 138 behavioral tasks. 139 140 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Surgical procedures 141 Rats were anesthetized with isoflurane (1-3.5% in oxygen) and injected with atropine (0.06 142 mg/mL). Eye ointment was applied to the eyes and was reapplied periodically throughout the surgery. 143 Once the pedal reflex was not displayed, their head was shaved and they were placed into a stereotaxic 144 instrument (Kopf). The incision site was sterilized with chlorhexidine solution and injected with lidocaine. 145 Hydrogen peroxide was used to control bleeding after the incision. After the skull was leveled and bregma 146 was identified, a stereotaxically mounted drill was used to mark craniotomy coordinates for dorsal HPC 147 and mPFC. Craniotomies were +3.1 mm anterior and +1.0 mm lateral to bregma (targeting prelimbic 148 cortex) and -3.7 mm posterior and +2.2 mm lateral to bregma (targeting dorsal CA1). A cerebellum 149

drill hole was made 12 mm posterior and -2.2 mm lateral to bregma. 4 bone screws (Fine 150 Science Tools) were inserted for stability and an additional bone screw was inserted above the 151 cerebellum for grounding. The mPFC wire bundle (2 stainless steel wires; wire diameter: 0.2 mm) was 152 implanted 2.6 mm ventrally at an 8-degree angle. A bundle of 4 wires (each wire staggered by 0.25 mm) 153 was implanted 2.5 mm ventrally at the HPC coordinates. The cerebellum reference wires (2 wires twisted 154 together) were implanted 1 mm ventrally. Wires were stabilized to the skull with Metabond. Dental acrylic 155 (Lang Dental) was used to secure a rod attached to an electrode interface board to the skull and to 156 stabilize the wire bundles. A copper mesh cage was placed around the drive components, and a wire 157 attached to the grounding screw was soldered to the cage and linked to the electrode interface board with 158 a gold pin. All other wires were also linked to the electrode interface board and liquid electrical tape was 159 applied over exposed wire. To protect drive components, a small weighing boat was velcroed on top of 160 the copper mesh cage and the implant was wrapped in a self-adhesive bandage. Neosporin and lidocaine 161 were applied to the skin surrounding the copper mesh. At the end of surgery, rats were injected with 162 flunixin (Banamine; 50 mg/mL) for post-surgery analgesia. In addition, 25 mL child’s ibuprofen (100 mg/5 163 dL) was added to the drinking water in the home cage. Rats completed a minimum of 1 week of recovery 164 before starting pre-training. 165 166 Pre-training 167 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint During goal box training, rats were trained to eat chocolate sprinkles from the weighing boats in 168 the goal zones of the maze. Wooden barriers were placed on both sides of the goal zone. Over 6 169 alternating trials, rats were placed in the left or right zone until they ate all the sprinkles or 3 minutes had 170 passed. Rats were required to eat all the sprinkles in under 90 seconds during each trial over two 171 consecutive days. 172 Forced run training familiarized rats with the T-maze route. Wooden barriers blocked the entry to 173 the stem of the maze and either the left or right goal arm at the start of each trial. Once the barrier at the 174 start box was lifted, rats traveled down the stem of the maze to the T-intersection and then proceeded 175 down the unblocked goal arm. Rats ate the reward in the goal zone and returned to the start box via the 176 return arm. A wooden barrier was then placed at the entry to the maze. Each session consisted of 12 177 trials (6 left and right in a random order). Rats spent 3-5 sessions completing the task until they 178 performed trials without guidance from the experimenter. Before continuing training, rats were acclimated 179 to performing the task while plugged in to the recording headstage. 180 181 Experimental design for behavioral tasks 182 The continuous alternation (CA) task is an HPC-independent task (Ainge et al., 2007) that follows 183 a spatial alternation rule (Figure 1B). To receive a reward, rats alternated between the left and right goal 184 arms over trials without returning to the start box. Rats were required to reach a criterion of 80% choice 185 accuracy (at least 32/40 trials correct) for two consecutive sessions. 186 Rats then began testing on the HPC-dependent delayed alternation (DA) task (Ainge et al., 2007; 187 Figure 1C). Rats were rewarded for alternating left and right goal arms over trials and returned to the start 188 box between trials to complete a delay. We systematically altered working memory load by changing the 189 delay duration between trials (10, 30, or 60 seconds). Each DA task session consisted of 36 delay trials 190 (plus an initial trial that rewarded rats for choosing either arm), with 12 trials of each delay length 191 pseudorandomly interleaved within the session. LFPs were recorded from the mPFC and HPC during the 192 task. Rats completed between 9-23 recording sessions. 193 194 Perfusion and histology 195 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Rats were anesthetized with isoflurane and were intraperitoneally injected with a veterinarian-196 approved mixture of xylazine and ketamine. Once rats no longer displayed the pedal and blink reflexes, 197 they were transcardially perfused with 100 mL of heparinized 0.1 M phosphate buffered saline (PBS) 198 followed by 100 mL of 4% paraformaldehyde in 0.1 M PBS (pH= 7.20). After the head was postfixed in 199 4% paraformaldehyde solution for 48 hours, the brain was extracted and transferred through 3 solutions 200 of 30% sucrose in 4% formaldehyde (24-72 hours in each solution until the brain sank) and stored at 4°C 201 until cryosectioning. A Leica cryostat (-20°C) was used to section brains in the coronal plane at 40 µm 202 and sections were stored in rostro-caudal order in a sucrose/ethylene glycol cryoprotectant solution at -203 20°C to verify electrode position. Electrode placement was verified by superimposing coronal section 204 images on a plate from the Paxinos and Watson (2006) stereotaxic atlas. 205 206 Video tracking and electrophysiology recordings 207 Video tracking data were obtained with a camera mounted to the ceiling that recorded LED lights 208 attached to the rat’s headstage at 30 Hz (Cheetah). Video tracking data from the DA task were visually 209 examined. Trials were excluded from analysis if they contained >10% tracking error in the stem entry to 210 choice point exit portion of the maze or had a failed stem entry/choice point exit (i.e. video tracking lost 211 the rat at these locations). If a trial contained a failed start box entry (when the rat returned for a delay at 212 the end of a trial), the following trial was removed. 213 A 64-channel digital recording system (Digital Lynx; Neuralynx) was used to record mPFC and 214 HPC LFPs, which were sampled at 2 kHz and filtered between 1-600 Hz using Cheetah software 215 (Neuralynx). LFPs were examined for artifacts and corresponding trials were excluded from analysis. 216 217 Behavioral analysis 218 Separating trials by delay length 219 11 AE and 7 SI rats were implanted with recording drives with LEDs on the headstage for video 220 tracking. To examine the effect of delay length on choice accuracy and VTEs, video tracking data were 221 used to calculate the time spent in the start box between trials. Trials were excluded from analysis if rats 222 did not leave the start box before the start of the following delay interval (e.g., a 10-second delay trial 223 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint where a rat did not exit the start box until after an actual delay of 30 seconds or greater had passed). Any 224 60-second delay trial with an actual delay above 100 seconds was excluded. Trials initiated more than 5 225 seconds before the intended delay time were also excluded (e.g., a 60-second delay trial where the trial 226 was initiated early, and the actual delay was less than 55 seconds). This step accounted for potential 227 disturbances in the testing room, such as the drive unplugging. 228 229 VTE trial identification 230 VTEs were identified using the integrated absolute change in angular velocity (IdPhi), a metric 231 that captures head movement complexity (Papale et al., 2012). Low IdPhi scores reflect direct paths 232 through the maze, whereas high IdPhi scores reflect pausing, reorienting, and head-sweeping behaviors 233 characteristic of VTEs. First, x and y position data from the stem to the choice point exit of the maze were 234 smoothed (smoothdata.m) using a moving average with a gaussian window (window size= 30; 1 second 235 of data). A discrete-time adaptive windowing method was used to calculate velocity in the x and y 236 dimensions (Janabi-Sharifi et al., 2000). The arctangent of the dX and dY components was taken and 237 unwrapped to determine the orientation of motion, Phi. The change in orientation, dPhi, was calculated by 238 applying the discrete-time adaptive windowing method to Phi. The integral of the absolute change in 239 orientation (|dPhi|) was calculated to obtain an IdPhi score for each trial. The natural log of IdPhi was 240 taken and lnIdPhi scores were z scored by rat. zlnIdPhi scores from AE and SI rats’ trials were shuffled 241 before examination of the data to blind the experimenter to group. 242 The VTE threshold is the value where the distribution of IdPhi scores deviates from a normal 243 distribution; this can be visualized as a “tail” off the right side of the distribution (Redish 2016, Figure 2A). 244 Trials with scores above this threshold typically represent VTE trials, whereas scores below this threshold 245 typically represent non-VTE trials (an example non-VTE trial is shown in the inset of Figure 2A). As the 246 deflection point occurred at a zlnIdPhi of 0.3, this value was selected as the VTE threshold, which is 247 similar to previously reported thresholds at the choice point (George et al., 2023). All trials with zlnIdPhi 248 scores above 0.3 were examined for verification as VTEs. Using the first visualization method (Figure 2B 249 left), position data from the stem to the choice point exit were plotted with the normalized velocity 250 overlaid. Trials with clear head-sweeping or pausing behavior at the T-intersection were retained as 251 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint VTEs. Trials with ballistic choice trajectories and/or complex head movements occurring before the choice 252 point entry or after the rat had entered a goal arm were marked as false positive VTE trials. Trials that 253 failed the first inspection were selected for a second round of visualization, when position data were 254 sequentially plotted to “play back" the selected trial. Trials that passed both visualization steps were 255 retained as VTE trials. A second method (Figure 2B right) was used to identify VTE trials with zlnIdPhi 256 scores below 0.3, where high velocity head-sweeping movements could have resulted in a below-257 threshold zlnIdPhi score and an incorrect classification as a non-VTE trial. This approach determined 258 instances when the rat entered rectangles in both the left and right goal arms of the T-maze during the 259 same trial. These trials were inspected to confirm head-sweeping behaviors at the choice point. Trials that 260 passed this inspection were classified as VTE trials. 261 We also examined VTEs in the T-maze stem. zlnIdPhi of 1.5 was chosen as the threshold value 262 based on the zlnIdPhi distribution generated using stem tracking data from each trial. Trials with above 263 threshold zlnIdPhi scores underwent visualization through Method 1. To examine VTEs at the choice 264 point during the CA task, data underwent both visualization methods, except lnIdPhi scores were not z-265 scored per rat as there were fewer trials. An lnIdPhi of 4.0 was determined to be the VTE threshold for the 266 CA task. 267 Analyses examining the proportion of VTEs per session (Figure 4A-B) included data up until 268 session 13, as each recording day contained data from at least half of the rats in each group until this 269 session. 270 271 DA task choice accuracy across sessions 272 To examine DA task choice accuracy over testing sessions, 12 implanted rats (7 AE, 5 SI) from 273 the current study were added to an additional dataset consisting of 39 rats (17 AE, 22 SI). These 274 additional rats completed the same experimental procedure as the rats from the current study except that 275 they were not implanted with recording drives. As rats in the previous dataset completed 6 sessions of DA 276 task testing, we analyzed task performance over these sessions in both groups. The sample size 277 accounts for rats excluded from choice accuracy analysis: 5 implanted rats were removed due to 278 recording issues that prevented at least 1 of the first 6 sessions from being completed, 1 implanted rat 279 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint was determined to be an outlier (greater than 3 scaled median absolute deviations from the median; 280 indicated by a red “X” in Figure 1E; this rat was excluded from all analyses) and 2 rats from the additional 281 dataset were found to have BAC results below 100 mg/dL and were excluded. If the recording headstage 282 became unplugged from implanted rats, the corresponding trial was excluded from the calculation of a 283 choice accuracy score for that session. 284 285 Perseverative errors 286 A perseverative error occurred if a rat made an incorrect choice on two consecutive trials of the 287 DA task (ex. left-right-right-right corresponds to correct-error1-error2). The proportion of perseverative 288 errors was calculated as the number of repeated choice errors divided by the total number of errors. 289 290 Electrophysiological analysis 291 Extracting LFPs in the choice point 292 LFPs were extracted over timestamps when rats occupied the choice point of the T-maze. The 293 3rd degree polynomial was removed from LFPs using detrend.m. The detrended signal was then z-294 scored to account for overall power distribution differences between rats due to increased signal 295 amplitude after copper mesh cages were introduced to the surgery procedure. 296 297 Coherence and power spectral density 298 To examine mPFC and HPC oscillatory activity and the magnitude of mPFC-HPC coupling during 299 choice point occupancy, power spectral density estimates (pwelch.m) and magnitude-squared coherence 300 (mscohere.m) were calculated over 1-50 Hz at a frequency resolution of 0.5 Hz. Power spectral density is 301 a measure of the power (squared amplitude) of a signal scaled by frequency. The log10 of the power 302 spectral density estimates was taken to account for 1/f noise. Magnitude-squared coherence is a metric 303 that describes the degree to which two signals are temporally correlated and ranges from 0 (no 304 correlation) to 1 (perfect correlation): 305 𝐶!" (𝑓) = '𝑃!"(𝑓)' # 𝑃!!(𝑓)𝑃""(𝑓) 306 307 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint The magnitude-squared coherence (Cxy) at a specified frequency (f) is the square of the absolute value of 308 the cross-power spectral density (Pxy) scaled by the power spectral density of each signal (Pxx, Pyy). As 309 1.25 seconds of data is sufficient for reliable estimates of theta coherence (Stout et al., 2023), trials that 310 did not reach this threshold were excluded from the analysis. To account for quick passes through the 311 choice point on non-VTE trials, LFPs were concatenated by session for non-VTE LFP analysis. 312 A moving window approach was used to examine the prevalence of mPFC-HPC coupling during 313 choice point occupancy on VTE and non-VTE trials. First, LFP signals were concatenated by rat. 314 Magnitude-squared coherence was then calculated from 6-10 Hz at a frequency resolution of 0.5 Hz over 315 1.25-second time windows (“coherence events”) that were gradually shifted by 250 milliseconds (Stout et 316 al., 2023; Figure 7A). The final samples of each rat’s concatenated signal were excluded as the remaining 317 data samples did not meet the 1.25-second minimum required for inclusion in coherence analysis. The 318 mean scores from each 1.25-second coherence event were compiled into empirical cumulative 319 distribution function (CDF) plots. 320 321 Machine learning analysis 322 To determine if our data could be used to predict whether a rat belonged to the AE or SI group, 323 we built 2 machine learning algorithms (K-Nearest Neighbors (KNN) Classifier and Euclidean Classifier) 324 using leave-one-out approaches. Features were z-scored to account for scaling differences. 325 In each iteration using the KNN Classifier, the Euclidean distance of the test data vector 326 (representing one rat) to each vector in the training data (representing every other rat) was calculated and 327 sorted. The 7 nearest vectors (neighbors) were determined (refer to Figure 8A), and the test data was 328 classified as belonging to the group to which at least 4 of 7 of the nearest neighboring rats belonged. To 329 determine if our classifier was performing above chance levels, we tested the classifier 1,000 separate 330 times using shuffled labels of AE and SI rats. A z-test was performed to test if the accuracy distribution 331 generated using the shuffled labels was significantly different from the accuracy score using the actual 332 labels (Sangiamo et al., 2020). In each iteration using the Euclidean Classifier, a vector representing all 333 the data from one rat was removed (test data). The remaining data (training data) were separated by 334 group, and the mean vectors were calculated. The Euclidean distance between the test data and each of 335 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint the mean vectors was determined, and the test data was then classified as belonging to the group that 336 corresponded to the shortest distance. Accuracy was calculated as the number of correct classifications 337 divided by the total number of iterations (17; each rat was excluded once). 338 339 Statistical analysis 340 All VTE choice accuracy analyses required a contribution of at least three trials at each level of 341 the independent variable (George et al., 2023). If a rat did not meet this parameter, the rat was excluded 342 from that test. Statistical analysis was conducted in MATLAB or JASP (ANOVAs). Significant ANOVA 343

(p<0.05) underwent Bonferroni correction for multiple comparisons. Corrected p values will be 344 referred to as pbonf. Information regarding statistical tests is stated in each result section. Cohen’s D was 345 calculated with computeCohen_D.m by R.G. Bettinardi (MATLAB) or in JASP. Figures were generated in 346 MATLAB and edited in Adobe Illustrator. 347 348 Code Accessibility 349 Data and code will be made available upon request. 350 351

352 Alcohol exposure disrupts choice behaviors 353 Despite previous reports of impaired executive functioning in our FASD rodent model (Gursky et 354 al., 2021) and impaired spatial working memory in other models of 3rd trimester AE (Thomas et al., 1996, 355 Wozniak et al., 2004), we did not observe a spatial working memory deficit as DA task accuracy did not 356 differ between groups (group: F(1,15)=0.512, p=0.485; delay by group: F(2,30)=0.140, p=0.870; repeated 357 measures ANOVA; N=7 SI rats, 10 AE rats; Figure 1D). The proportion of correct trials decreased with 358 increasing delay in both groups (F(2,30)=42.376, p<0.001, η2p=0.739; post hoc comparisons: 10-30s 359 t=2.806, pbonf=0.026, d=0.758; 10-60s t=8.996, pbonf<0.001, d=2.429; 30-60s t=6.190, pbonf<0.001, 360 d=1.671; two-sample, two-tailed t-test). While spatial working memory was not disrupted by AE, we found 361 that that the AE group spent significantly less time in the choice point than SI controls (t(15)=2.528, 362 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint p=0.023, d=1.246, two-sample, two-tailed t-test; N= 7 SI rats, 10 AE rats; Figure 1E). Together, these 363

indicate that AE altered choice behaviors without disrupting spatial working memory. 364 365 Alcohol exposed rats engage in less vicarious trial and errors than controls on the delayed 366 alternation task 367 To further characterize how choice behaviors were impacted by AE, we investigated VTEs, which 368 are behaviors associated with flexible decision-making, deliberation, and uncertainty (George et al., 2023; 369 Papale et al., 2012; Redish, 2016; Schmidt et al., 2013). We first examined whether there was a 370 relationship between the proportion of trials with a VTE, working memory demand, and AE (Figure 2C; 371 N=7 SI rats, 10 AE rats). We found a main effect of group on the proportion of trials that had VTEs, with 372 the AE group exhibiting a lower proportion of VTE trials than SI controls (F(1,15)=8.540, p=0.011, 373 η2p=0.363; repeated measures ANOVA). There was no main effect of delay length or delay by group 374 interaction on the proportion of VTE trials, demonstrating that working memory load did not affect overall 375 VTE occurrence (delay: F(2,30)=2.147, p=0.134; delay by group: F(2,30)=0.319, p=0.729). 376 While examining tracking data to confirm VTEs at the choice point, we noticed instances of rats 377 displaying VTE-like behaviors on the maze stem (Figure 2D right). We were curious if these “stem VTEs” 378 would also be lower in the AE group compared to the SI group. A repeated measures ANOVA revealed a 379 main effect of group on VTE trial proportion in the stem of the maze, with the AE group showing a lower 380 proportion of trials with stem VTEs than the SI group (F(1,15)=4.583, p=0.049, η2p=0.234; N=7 SI rats, 10 381 AE rats; Figure 2D left). There was no effect of delay length or delay by group interaction on VTE 382 proportion in the T-maze stem (delay: F(2,30)=2.995, p=0.065; delay by group: F(2,30)=0.907, p=0.415). 383 Both the SI and AE groups performed a greater proportion of VTEs in the choice point than the stem of 384 the maze (SI: t(6)=8.182, p=0.0002, d=3.093; AE: t(9)=4.581, p=0.001, d=1.449; one-sample, two-tailed t-385 test against a null of 0; data not shown). 386 Our findings suggest that developmental AE leads to less deliberation during decision-making in 387 adulthood. However, an alternative explanation is that our results instead reflect a motor impairment 388 (Goodlett et al., 1991; Klintsova et al., 1998; Thomas et al., 1996), as AE during the brain growth spurt 389 also damages the cerebellum (Bonthius & West, 1991; Hamre & West, 1993). To investigate this 390 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint possibility, we examined VTEs at the choice point during the CA task, a task with a comparatively low 391 working memory demand compared to the DA task. Tracking data were recorded from 8 AE rats and 7 SI 392 rats during 1-5 CA task sessions occurring late in training. In contrast to the DA task, there was no 393 difference in time spent in the choice point or the proportion of trials with VTEs between groups on the CA 394 task (time spent: t(13)=0.062, p=0.951, data not shown; VTE: t(13)=0.556, p=0.587; two-sample, two-395 tailed t-test; Figure 2G). As the AE group was capable of performing VTEs at similar levels as the SI 396 group, it is unlikely that motor impairments explain VTE differences on the DA task. 397 We next investigated whether choice accuracy on VTE trials differed between groups and if delay 398 length affected performance on these trials. In contrast to our overall DA task accuracy results, we found 399 that choice accuracy did not change across delays on trials with VTEs at the choice point (F(2,24)=1.531, 400 p=0.237; repeated measures ANOVA; N=7 SI rats, 7 AE rats; Figure 2E). AE and SI groups also 401 performed similarly on choice point VTE trials across delays (group: F(1,12)=0.007, p=0.933; delay by 402 group: F(2,24)=0.236, p=0.792). Due to the low trial count of stem VTEs, we did not analyze the 403 relationship between choice accuracy and delay length. 404 As VTEs are associated with uncertainty and conflict, they are also related with poorer task 405 performance compared to non-VTE trials (Amemiya & Redish, 2016). We examined whether this 406 relationship was disrupted after AE and how VTE location in the maze (either the stem or the choice 407 point) impacted choice accuracy (Figure 2F). Both non-VTE trials and stem VTE trials showed higher 408 choice accuracy than choice point VTE trials (F(2,26)=10.105, p<0.001, η2p=0.437; repeated measures 409 ANOVA; post hoc comparisons: non-VTE vs choice point VTE t=4.449, pbonf<0.001, d=1.387; stem VTE 410 vs choice point VTE t=2.782, pbonf=0.030, d=0.867; two-sample, two-tailed t-test; N= 7 SI rats, 8 AE rats). 411 Interestingly, choice accuracy on stem VTE trials was not significantly different from choice accuracy on 412 non-VTE trials (t=1.667, pbonf=0.323; two-sample, two-tailed t-test). Choice accuracy was not affected by 413 AE (group: F(1,13)=1.139, p=0.305; trial type by group: F(2,26)=0.438, p=0.650). 414 Together, our results suggest that AE during the brain growth spurt leads to reduced deliberative 415 behaviors during HPC-dependent working memory, as the AE group exhibited fewer VTEs on the DA task 416 while groups showed similar amounts of VTEs on the CA task. While VTE frequency was lowered after 417 AE, the AE group did not show a choice impairment on VTE trials. We also found that VTEs were not 418 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint limited to locations near the T-intersection of the maze, demonstrating that rats occasionally began 419 engaging in these behaviors shortly after trial initiation. Moreover, engaging in deliberation early in the 420 trial (in the stem versus the choice point) may have benefited impending choice accuracy. Choice 421 accuracy on choice point VTEs was not affected by delay, indicating that these behaviors manifested 422 similarly regardless of working memory load. 423 424 Disturbed relationship between vicarious trial and error and choice outcomes following alcohol 425 exposure 426 VTEs have been shown to be more common on error trials compared to correct trials (Bett et al., 427 2012; Schmidt et al., 2013; but see Miles et al., 2024). To investigate the relationship between AE, trial 428 accuracy, and delay duration on choice point VTE behaviors, we compared the proportion of VTEs 429 occurring on correct and error trials (Figure 3A) for the 10-, 30-, and 60-second delays in the AE and SI 430 groups (N= 7 SI rats, 10 AE rats). There was no significant 3-way interaction between AE, trial accuracy, 431 and delay (F(2,30)=1.167, p=0.325; repeated measures ANOVA). However, there was a significant 432 interaction between trial accuracy and group, as the proportion of VTE error trials (Figure 3B), but not 433 VTE correct trials (Figure 3C), was lower in the AE group compared to the SI group (trial accuracy by 434 group: F(1,15)=11.316, p=0.004, η2p=0.430; trial accuracy: F(1,15)=70.124, p<0.001, η2p=0.824; post hoc 435 comparisons: error AE vs error SI t=-4.730, pbonf<0.001, d=1.886; correct AE vs correct SI t=-1.777, 436 pbonf=0.537; two-sample, two-tailed t-test). Both groups also performed a greater proportion of VTE error 437 trials than VTE correct trials (correct AE vs error AE t= -3.904, pbonf=0.008, d=0.877; correct SI vs error SI 438 t=-7.652, p<0.001, d=2.054; two-sample, two-tailed t-test). 439 There was also a significant trial accuracy by delay interaction, with the proportion of VTE error 440 trials decreasing with delay duration and the greatest proportion occurring on 10-second delay trials (trial 441 accuracy by delay: F(2,30)=16.510, p<0.001, η2p=0.524; delay F(2,30)=14.375, p<0.001, η2p=0.489; 442 repeated measures ANOVA; post hoc comparisons: 10s error-30s error t=5.977, pbonf<0.001, d=1.499; 443 10s error-60s error t=7.314, pbonf<0.001, d=1.834; 30s error-60s error t=1.336, pbonf=1.000; two-sample, 444 two-tailed t-test). Therefore, VTE error trials followed an opposite trend to error patterns typically 445 observed in delayed alternation tasks, which increase with delay duration, as reported in our dataset (See 446 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Figure 1D) and previous studies that did not separate VTE from non-VTE trials (Ainge et al., 2007; de 447 Mooij-van Malsen et al., 2023; Layfield et al., 2015). In contrast, there was no relationship between the 448 proportion of correct trials with VTEs and delay duration (10s correct-30s correct t=0.464, pbonf=1.000; 10s 449 correct-60s correct t=0.428, pbonf=1.00, 30s correct-60s correct t=-0.036, pbonf=1.000). 450 Our results indicate that the lower proportion of VTEs exhibited by the AE group (Figure 2C) is 451 likely driven by a reduction in VTEs performed during error trials compared to the SI group. Furthermore, 452 while rats made fewer choice errors on 10-second delay trials compared to 30- and 60-second trials, a 453 higher proportion of these trials had VTEs. 454 455 Altered relationship between experience and vicarious trial and error in the alcohol exposed 456 group 457 We were next interested in examining if VTE differences between groups were associated with 458 choice accuracy differences at the session level on the DA task. As VTEs are inversely related to learning 459 (Griesbach et al., 1998; Hu & Amsel, 1995; Muenzinger, 1938; Tolman, 1939), we first predicted that the 460 proportion of VTE trials per session would be negatively correlated with choice accuracy. Consistent with 461 previous findings, VTE proportion was negatively correlated with accuracy for both groups (SI: r=-0.3562, 462 p=0.0015; AE: r=-0.3467, p=0.0004; r=correlation coefficient, Pearson’s correlation; N=77 sessions from 463 SI rats, 99 sessions from AE rats; Figure 4A). We also predicted that the greatest proportion of VTEs 464 would occur during the first DA task sessions when rats would need to adjust their strategy to address 465 changes in task demands relative to the CA task and that these behaviors would decrease over sessions. 466 Interestingly, while the SI group demonstrated a reduction in VTE proportion over sessions, this trend was 467 not observed in the AE group, which showed no change in VTE proportion over sessions (SI: r=-0.3806, 468 p=0.0006; AE: r=-0.0569, p=0.576; Pearson’s correlation; N=77 sessions from SI rats, 99 sessions from 469 AE rats; Figure 4B). 470 Given that the proportion of VTE trials was lower in the AE group compared to the SI group and 471 the frequency of VTE trials did not change with experience in the AE group, we predicted that the AE 472 group would show an impairment on the task over sessions. We included rats from a previous dataset 473 that completed the same experimental procedure except DA task testing stopped after session 6 and 474 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint recording drives were not implanted (combined N= 27 SI rats, 24 AE rats). A repeated measures ANOVA 475 revealed that there was no interaction between group and session and no effect of group on choice 476 accuracy (group by session: F(5,245)=1.664, p=0.144; group: F(1,49)=0.008, p<0.929; Figure 4C). There 477 was a main effect of session on choice accuracy (F(5,245)=7.665, p<0.001, η2p=0.135). Both SI and AE 478 groups improved across sessions (SI: r=0.3776; p<0.001 AE: r=0.1807, p=0.030; Pearson’s correlation). 479 Together, these results further confirm that although VTE behaviors were disrupted in AE rats, this 480 disruption did not prevent rats from successfully performing and improving on the DA task. 481 482 The functionality of deliberative behaviors is reduced after alcohol exposure 483 Reorienting behaviors have previously been shown to enhance future decision-making (George 484 et al., 2023). As there was a disturbed relationship between VTEs and performance over sessions in the 485 AE group, we were next interested in determining whether the relationship between VTEs and 486 subsequent performance was also altered. We examined choice accuracy on the trial following a VTE trial 487 and found that the AE group had lower choice accuracy following 10-second delay trials with VTEs 488 compared to the SI group (t(12)=2.508, p=0.028, d=1.295; two-sample, two-tailed t-test; N=7 SI rats, 7 AE 489 rats; Figure 5A). This relationship did not exist when considering non-VTE 10-second delay trials 490 t(15)=0.930, p=0.367; N=7 SI rats, 10 AE rats; Figure 5D). Therefore, the impaired performance of the AE 491 group following 10-second delay trials was not a general characteristic of performance and was specific 492 to trials following VTE trials. In contrast, both groups performed similarly on the trial following 30-second 493 and 60-second delay trials with VTEs (30s: t(12)=-1.016, p=0.330; 60s: t(12)=-1.632, p=0.129; Figure 5B-494 C). Due to the trial sequence of the DA task, trials following 10-, 30-, and 60-second trials were not evenly 495 distributed (Figure 5E). However, these differences do not explain the impaired performance of the AE 496 group after 10-second VTE trials compared to SI controls, as both groups had similar distributions of 497 delay trials following each type of VTE trial. 498 As these results indicated that flexibility may be impaired in AE rats, we decided to investigate 499 measures of executive dysfunction. Inactivation of the mPFC (G.-W. Wang & Cai, 2006), Re (Stout et al., 500 2022; Viena et al., 2018), and HPC (Hallock et al., 2013) is associated with choice inflexibility, reflected 501 as an increase in repeated choice errors, known as perseverative errors. Similarly, rodent models of third 502 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint trimester AE have shown increased perseverative errors during spatial working memory and serial spatial 503 discrimination reversal tasks (Thomas et al., 1996, 1997). These findings posed the possibility that we 504 may see an increase in inflexible choice behaviors in our 3rd trimester FASD rodent model. However, we 505 found that the AE and SI groups engaged in a similar proportion of perseverative errors during the DA 506 task (t(15)=0.175, p=0.864; two-sample, two-tailed t-test; N= 7 SI rats, 10 AE rats; Figure 5F). 507 Given previous work has shown that nucleus reuniens inactivation increases VTEs during 508 perseverative error sequences (Stout et al., 2022), we decided to investigate the relationship between the 509 proportion of perseverative errors and the proportion of VTEs from each rat’s recording sessions. 510 Interestingly, perseverative errors were positively correlated with VTEs in the AE group only, suggesting 511 that AE altered performance such that flexible decision-making behaviors became associated with 512 inflexible decision-making behaviors (individual rats: SI (7 rats) r=-0.0201, p=0.9659; AE (10 rats) 513 r=0.6608, p=0.0375; individual sessions: SI (90 sessions) r=0.0779, p=0.4654; AE (121 sessions) 514 r=0.3894, p<0.001; Pearson’s correlation; Figure 5G). Collectively, these findings suggest that VTE 515 efficacy has been reduced in AE rats as they did not facilitate a flexible choice strategy as reflected in SI 516 controls. 517 518 Alcohol exposure alters mPFC theta oscillations and HPC beta oscillations 519 mPFC-HPC theta synchrony via the nucleus reuniens has been implicated in decision-making 520 (Hallock et al., 2016) and VTE behaviors (Stout et al., 2022). Therefore, we were interested in examining 521 the effects of AE on mPFC and HPC physiology and synchrony in the theta band (6-10 Hz) during VTEs 522 (Figure 6A). 7 AE and 4 SI rats were included in LFP analysis after verifying electrode placements. The 523 power spectral densities of mPFC and HPC LFPs recorded during choice point occupancy in both the AE 524 and SI groups are shown as a function of frequency in Figure 6B-C (left). We found that theta power in 525 the mPFC was significantly lower in the AE group compared to the SI group during VTEs (t(9)=2.534, 526 p=0.032, d=1.588; two-sample, two-tailed t-test; Figure 6B middle). To determine if this effect was specific 527 to VTEs, we next examined non-VTE trials. After outlier removal, we found that mPFC theta power was 528 also significantly lower in the AE group compared to the SI group during non-VTE trials (t(7)=2.716, 529 p=0.030; d=1.822; N=4 SI rats, 5 AE rats; Figure 6E left). Follow-up analysis revealed that the proportion 530 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint of VTE trials performed by rats in the AE group, but not the SI group, was negatively correlated with 531 mPFC theta power during VTE trials, but not non-VTE trials (VTE: AE r=0.7843, p=0.0368; SI r=0.4094, 532 p=0.5906; Pearson’s correlation; Figure 6H; non-VTE: AE r=-0.2895, p=0.6366; SI: r=0.1588, p=0.8412; 533 data not shown). In contrast, HPC theta power and mPFC-HPC theta coherence were not different 534 between groups during VTEs and non-VTEs (VTE power: t(9)=0.291, p=0.778; Figure 6C middle; VTE 535 coherence: t(9)=-0.232, p=0.822; Figure 6D middle; non-VTE power: t(9)=0.930, p=0.376; Figure 6F left; 536 non-VTE coherence: t(9)=0.227, p=0.826; Figure 6G left). 537 Beta rhythms (15-30 Hz) have also been associated with VTEs (Miles et al., 2024) and 538 synchronize in the mPFC-nucleus reuniens-HPC circuit during memory tasks (de Mooij-van Malsen et al., 539 2023; Jayachandran et al., 2022). We found that HPC beta power was significantly higher in the AE group 540 compared to the SI group during both VTE and non-VTE trials (VTE: t(9)=-2.520, p=0.033, d=1.580; 541 Figure 6C right; non-VTE: t(9)=-3.188, p=0.011; d=1.998; Figure 6F right). Conversely, mPFC beta power 542 and mPFC-HPC beta coherence during VTEs and non-VTEs were not significantly different between 543 groups (VTE power: t(9)=0.731, p=0.483, Figure 6B right; VTE coherence: t(9)=-0.497, p=0.631; Figure 544 6D right; non-VTE power: t(9)=-0.372, p=0.719; Figure 6E right; non-VTE coherence: t(9)=-1.024, 545 p=0.333; Figure 6G right). 546 Together, these results suggest that AE during the brain growth spurt alters mPFC theta rhythms 547 and HPC beta rhythms during both VTEs and non-VTEs without disrupting the magnitude of mPFC-HPC 548 synchrony. 549 550 mPFC-HPC theta coupling events are less common after alcohol exposure 551 Our results suggested that the magnitude of mPFC-HPC theta synchrony during decision-making 552 was not different between groups. It remained possible that AE could disturb the commonality of mPFC-553 HPC coupling events, rather than the magnitude. For example, magnitude coherence measures over 554 choice point occupancy could have masked differences in how frequently the mPFC and HPC 555 synchronized over shorter timescales. Therefore, we next used a moving window approach to calculate 556 mPFC-HPC theta coherence over 1.25-second “coherence events” (refer to Methods; example trials with 557 similar magnitude coherence and different coherence event distributions are shown in Figure 7A). We first 558 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint validated this approach by replicating our previous magnitude coherence results from Figure 6D using the 559 mean coherence magnitude across events for each rat (t(9)=1.129, p=0.288; two-sample, two-tailed t-560 test; N= 4 SI rats, 7 AE rats; Figure 7B). 561 Interestingly, whereas magnitude coherence was not different between groups using either 562 approach, the distributions of theta coherence events were significantly different between groups 563 (k=0.124, p<0.001; two-sample Kolmogorov-Smirnov Test; Figure 7C). The coherence event distribution 564 of the AE group was shifted leftward compared to the SI group, suggesting that AE led to less frequent 565 mPFC-HPC theta coupling. This effect was not specific to VTE trials, as we also observed differences 566 between groups in the distributions of theta coherence events during non-VTE trials (k=0.148, p<0.001; 567 Figure 7D). We noticed variability in the number of coherence events that each AE rat contributed to the 568 overall coherence event distribution. To confirm that these effects could also be observed at the rat level, 569 we then collapsed theta coherence events across VTE and non-VTE trials for each rat and tested the 570 distributions of each AE rat against the SI group distribution. We found that 6/7 AE rats showed 571 coherence event distributions that were significantly different from the SI group distribution (rat 1: k=0.216 572 ; p<0.001; rat 2: k=0.083; p<.001; rat 3: k=0.079; p=0.0013; rat 4: k=0.335; p<0.001; rat 5: k=0.016; 573 p=0.462; rat 6: k=0.046; p=0.025; rat 7: k=0.141; p<0.001; Figure 7E). Furthermore, the majority of AE 574 rats demonstrated leftward-shifted distributions, indicating that mPFC-HPC theta coupling events were 575 less common compared to the SI group. Collectively, these results indicate that the incidence of mPFC-576 HPC synchronous events, but not the magnitude of synchrony, is altered after AE and that this alteration 577 is not specific to VTE trials. 578 579 Using machine learning to predict treatment of alcohol exposed and sham intubated rats 580 We were next interested in determining whether we could predict the treatment of each rat (as AE 581 or SI) using machine learning. Features consisted of the following categories: time spent in choice point, 582 proportion of VTE trials in the choice point by delay, proportion of VTE trials in the stem by delay, 583 proportion of VTE error trials in the choice point by delay, proportion of VTE correct trials in the choice 584 point by delay, proportion of perseverative error trials, and proportion correct by delay (all results included 585 a full sample size for analysis of 10 AE rats and 7 SI rats, therefore LFP data was excluded). We built a 586 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint KNN classifier with K=7, as this value resulted in the highest accuracy while considering the fewest 587 neighbors (Figure 8A). We found that postnatal treatment as AE or SI was predicted with above-chance 588 accuracy, correctly classifying 9/10 AE rats and 4/7 SI rats (overall accuracy 76.5%; z=2.393, p=0.017; 589 one-sample, two-tailed z-test; Figure 8B). To further validate our KNN Classifier, we also built a Euclidean 590 Classifier to predict treatment. Similar to the KNN Classifier, the Euclidean Classifier correctly predicted 591 9/10 AE rats and 4/7 SI rats, performing at an identical accuracy of 76.5% (Figure 8C). These results 592 demonstrate our classifiers could reliably predict whether a rat was exposed to alcohol during 593 development based on behavioral data from the DA task. 594 To determine which combination of behaviors could characterize a potential phenotype for third 595 trimester AE in our model, we identified which categories were most important for the correct 596 classification of rats as AE or SI. We then iteratively removed each category from the KNN classifier, 597 determined accuracy, and found that classifier accuracy decreased below chance levels (p>0.05) only 598 when time spent in the choice point, the proportion of VTE error trials at each delay, or proportion correct 599 at each delay were excluded (Figure 8D). Using solely these three categories to predict treatment, the 600 classifier again performed at an accuracy of 76.5%, which was above chance levels (z=2.258, p=0.024; 601 Figure 8E). In contrast, when all remaining categories were used to predict treatment, the classifier 602 performed below chance levels (z=1.345, p=0.179). 603 These results demonstrate that only three categories were required for our classifier to reach 604 peak accuracy. In addition, while accuracy on the DA task was not significantly different between groups 605 (Figure 1D), its interaction with time spent in the choice point (Figure 1E) and the proportion of VTE error 606 trials (Figure 3B) was important in characterizing a phenotype of AE versus SI rats. 607 608

609 In this study, we show that AE during the brain growth spurt led to disrupted choice behaviors and 610 altered mPFC-HPC physiology and connectivity without impairing spatial working memory, suggesting a 611 selective disruption to executive function following AE. We further demonstrate that a machine learning 612 algorithm could predict whether rats were AE based on behavioral measures from our task, identifying a 613 phenotype for our model of third trimester AE. 614 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint The AE group performed a lower proportion of VTEs in the choice point and stem of the T-maze 615 compared to the SI group during the DA task. There was no difference in VTE proportion between groups 616 on the CA task, showing that AE rats can perform VTEs normally during a task that has a low working 617 memory demand and does not require HPC (Ainge et al. 2007). In contrast, the DA task has a working 618 memory component that increases with delay duration, is HPC-dependent (Ainge et al. 2007), and relies 619 on mPFC-HPC interactions via the nucleus reuniens (Hallock et al., 2016), particularly during VTEs (Stout 620 et al., 2022). Consequently, the lower proportion of VTEs in the AE group compared to the SI group is 621 likely due to AE-related dysfunction within this circuit disrupting processes underlying deliberation. 622 Although VTEs were reduced, spatial working memory was unimpaired in the AE group. This 623

was surprising given reductions in VTEs are associated with learning and memory deficits (Bett et 624 al., 2012; Blumenthal et al., 2011; Griesbach et al., 1998; Hu & Amsel, 1995; Kidder et al., 2021). VTEs 625 were also unrelated to task acquisition in the AE group, indicating that the AE group utilized a strategy 626 that relied less on VTEs but was effective in making correct choices. In contrast, in agreement with 627 previous studies, the SI group showed a reduction in VTEs across sessions that coincided with increased 628 choice accuracy, suggesting rats utilized a deliberative strategy upon task introduction that became less 629 necessary as proficiency increased (Griesbach et al., 1998; Hu & Amsel, 1995; Muenzinger, 1938; 630 Redish, 2016; Tolman, 1939). To our knowledge, we are the first to show that VTE frequency and 631 accuracy are unaffected by working memory demand, which likely relates to VTEs reflecting uncertainty 632 (Amemiya & Redish, 2016; Schmidt et al., 2013). We also demonstrate that VTEs in the T-maze stem had 633 higher accuracy than VTEs in the choice point, indicating that the timing of VTEs relative to the choice 634 has implications for subsequent decision-making. 635 The AE group performed fewer VTE error trials than the SI group, yet both groups committed a 636 similar proportion of choice errors. These results reveal a fundamental difference in choice behavior 637 during error trials following AE. This prediction is supported by our finding that removing VTE error trials 638 as a category from our KNN Classifier resulted in the greatest decrease in classifier accuracy. As VTEs 639 are thought to reflect the evaluation of choices during indecision (Redish, 2016), our results suggest that 640 the AE group was less likely to engage in deliberative behaviors when uncertain. However, while VTEs 641 were disrupted in the AE group, these behaviors appeared to reflect similar processes in both groups. For 642 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint example, VTEs were associated with error trials (Bett et al., 2012; Schmidt et al., 2013) and had lower 643 accuracy than non-VTE trials (Amemiya & Redish, 2016). Sessions with high proportions of VTEs also 644 tended to have low choice accuracy (Griesbach et al., 1998; Hu & Amsel, 1995; Tolman, 1939). 645 Working memory and VTEs were related during choice errors. While a shorter inter-trial delay 646

in an easier version of the task, it also increases the potential for interference between trials. A 647 recent study found that reorienting behaviors similar to VTEs increased on the delayed non-match to 648 place task when rats were blocked from alternating on the sample phase of the current trial relative to the 649 choice phase of the preceding trial, even though the alternation rule was irrelevant during the sample 650 phase (George et al., 2023). In the current context, rats may have been unable to dissociate the previous 651 from the current trial after 10-second delays, and this conflict could have resulted in VTE occurrence and 652 choice error. As proactive interference decreases with increased inter-trial delay (Grant, 1981), 653 interference would have been less likely after 30- and 60-second delays. Collectively, errors after 10-654 second delays may be more reflective of interference rather than forgetfulness, whereas the latter may 655 play a larger role in errors after 30- and 60-second delays. We propose that VTEs after the 10-second 656 delay emerge in part due to this interference, whereas VTEs following 30- and 60-second delays arise 657 due to uncertainty. This prediction may explain the choice deficit following 10-second delay VTE trials, but 658 not 30- or 60-second delay VTE trials in the AE group, as VTEs performed to resolve interference rather 659 than deliberate choice options may have separate implications for upcoming behavior. 660 We also observed that perseverative errors and VTEs were positively correlated in the AE group. 661 The relationship between flexible (VTE) and inflexible (perseverative error) choice behaviors is 662 contradictory but agrees with previous findings of increased VTEs during perseverative error sequences 663 after nucleus reuniens inactivation (Stout et al., 2022). As the AE group did not demonstrate a greater 664 proportion of perseverative error sequences compared to controls, this altered relationship relates to a 665 reduction in the effectiveness of VTE behaviors rather than an increase in inflexible behaviors. mPFC-666 nucleus reuniens-HPC circuit dysfunction likely contributes to the dissociation between VTEs and flexible 667 decision-making in the AE group given that disrupting these regions affects both VTE and perseverative 668 error behaviors (G.-W. Wang & Cai, 2006; Hallock et al., 2013; Hu & Amsel, 1995; Kidder et al., 2021; 669 Stout et al., 2022; Viena et al., 2018). 670 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint In support of circuit disruption following AE, we found that mPFC theta rhythms and HPC beta 671 rhythms were altered during both VTE and non-VTE trials in the AE group compared to the SI group. 672 Theta and beta rhythms have been implicated in VTEs, as both are increased in the mPFC during VTE 673 trials compared to non-VTE trials (Miles et al., 2024) and theta is present in the HPC during VTEs 674 (Amemiya & Redish, 2016; Johnson & Redish, 2007). As disrupting the circuitry involved in VTEs has 675 been shown to alter mPFC and HPC physiology and affect VTE behavior (Schmidt et al., 2019; Stout et 676 al., 2022), changed oscillatory activity in the theta and beta ranges may contribute to altered VTE 677 functionality in our FASD model. This prediction is supported by our finding that mPFC theta power was 678 negatively correlated to the proportion of VTEs performed by rats in the AE group, linking mPFC 679 dysfunction to the observed VTE deficit. These neurophysiology results may further relate to executive 680 functioning deficits previously described in our rodent model (Gursky et al., 2021). 681 The prevalence of mPFC-HPC synchronous events was also altered in the AE group compared 682 to the SI group. Interestingly, whereas the magnitude of mPFC-HPC theta coherence was not different 683 between groups, our results indicate that mPFC-HPC theta coupling events were less common in the AE 684 group compared to the SI group. Altered mPFC-HPC theta coupling was not specific to VTE trials, 685 suggesting that these changes are a characteristic of mPFC-HPC functional connectivity after 686 developmental AE. It is possible that reorganization within the brain after AE conserved the magnitude of 687 mPFC-HPC synchrony, but not the incidence of synchronous events. We suspect these changes may 688 have conserved spatial working memory but disrupted aspects of decision-making, such as VTEs 689 becoming less effective for AE rats. 690 In further demonstration that AE during the brain growth spurt has robust effects on behavior in 691 adulthood, our KNN classifier was effective in identifying whether rats were AE using behavioral 692 measures from task performance. We found that time spent in the choice point, VTE error trials, and the 693 proportion of correct trials were most important in accurately classifying rats as belonging to the AE or SI 694 group, and therefore may be among the measures that characterize the FASD phenotype after third 695 trimester AE. 696 Our results suggest that the AE group occasionally attempted to deliberate. However, disruptions 697 to mPFC-HPC circuitry may have impaired the ability to engage in VTEs when rats performed a task that 698 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint relied on the integrity of this circuit. Moreover, as the AE group was sometimes unable to utilize these 699 behaviors to inform future decision-making, the benefit of VTEs as a flexible choice strategy was reduced, 700 which could have diminished the need to perform these behaviors as frequently as controls. These 701 factors could have promoted a strategy that did not require VTEs and spared working memory in the AE 702 group. 703 Collectively, these findings contribute to a better understanding of the effects of third trimester AE 704 on decision-making by providing evidence for behavioral disruptions and neurophysiological alterations 705 within the mPFC-HPC circuit that offer insight into executive functioning deficits after prenatal AE. These 706

further identify the mPFC-HPC network as a target for therapeutic interventions in FASD patients. 707 708 Author contributions 709 H.L.R. collected data, performed analysis, and wrote the manuscript. S.K. generated rats and collected 710 data. J.J.S. and A.L.G. provided analysis feedback. A.K. guided animal generation. A.K. and A.L.G. 711 acquired funding. All authors conceptualized questions and contributed to the writing of this manuscript. 712 713

714 Ainge, J. A., van der Meer, M. A. A., Langston, R. F., & Wood, E. R. (2007). Exploring the role of context-715 dependent hippocampal activity in spatial alternation behavior. Hippocampus, 17(10), 988–1002. 716 https://doi.org/10.1002/hipo.20301 717 Amemiya, S., & Redish, A. D. (2016). Manipulating Decisiveness in Decision Making: Effects of Clonidine 718 on Hippocampal Search Strategies. The Journal of Neuroscience, 36(3), 814. 719 https://doi.org/10.1523/JNEUROSCI.2595-15.2016 720 Benchenane, K., Peyrache, A., Khamassi, M., Tierney, P. L., Gioanni, Y., Battaglia, F. P., & Wiener, S. I. 721 (2010). Coherent Theta Oscillations and Reorganization of Spike Timing in the Hippocampal- 722 Prefrontal Network upon Learning. Neuron, 66(6), 921–936. 723 https://doi.org/10.1016/j.neuron.2010.05.013 724 Bett, D., Allison, E., Murdoch, L., Kaefer, K., Wood, E., & Dudchenko, P. (2012). The neural substrates of 725 deliberative decision making: Contrasting effects of hippocampus lesions on performance and 726 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint vicarious trial-and-error behavior in a spatial memory task and a visual discrimination task. 727 Frontiers in Behavioral Neuroscience, 6. 728 https://www.frontiersin.org/articles/10.3389/fnbeh.2012.00070 729 Blumenthal, A., Steiner, A., Seeland, K., & David Redish, A. (2011). Effects of pharmacological 730 manipulations of NMDA-receptors on deliberation in the Multiple-T task. Neurobiology of Learning 731 and Memory, 95(3), 376–384. https://doi.org/10.1016/j.nlm.2011.01.011 732 Bonthius, D. J., & West, J. R. (1991). Permanent neuronal deficits in rats exposed to alcohol during the 733 brain growth spurt. Teratology, 44(2), 147–163. https://doi.org/10.1002/tera.1420440203 734 Churchwell, J. C., & Kesner, R. P. (2011). Hippocampal-prefrontal dynamics in spatial working memory: 735 Interactions and independent parallel processing. Behavioural Brain Research, 225(2), 389–395. 736 https://doi.org/10.1016/j.bbr.2011.07.045 737 Coles, C. (1994). Critical Periods for Prenatal Alcohol Exposure: Evidence From Animal and Human 738 Studies. Alcohol Health and Research World, 18(1), 22–29. 739 de Mooij-van Malsen, J. G., Röhrdanz, N., Buschhoff, A.-S., Schiffelholz, T., Sigurdsson, T., & Wulff, P. 740 (2023). Task-specific oscillatory synchronization of prefrontal cortex, nucleus reuniens, and 741 hippocampus during working memory. iScience, 26(9), 107532. 742 https://doi.org/10.1016/j.isci.2023.107532 743 Dobbing, J., & Sands, J. (1979). Comparative aspects of the brain growth spurt. Early Human 744 Development, 3(1), 79–83. https://doi.org/10.1016/0378-3782(79)90022-7 745 Floresco, S. B., Seamans, J. K., & Phillips, A. G. (1997). Selective Roles for Hippocampal, Prefrontal 746 Cortical, and Ventral Striatal Circuits in Radial-Arm Maze Tasks With or Without a Delay. The 747 Journal of Neuroscience, 17(5), 1880. https://doi.org/10.1523/JNEUROSCI.17-05-01880.1997 748 George, A. E., Stout, J. J., & Griffin, A. L. (2023). Pausing and reorienting behaviors enhance the 749 performance of a spatial working memory task. Behavioural Brain Research, 446, 114410. 750 https://doi.org/10.1016/j.bbr.2023.114410 751 Grant, D. S. (1981). Intertrial interference in rat short-term memory. Journal of Experimental Psychology: 752 Animal Behavior Processes, 7(3), 217–227. https://doi.org/10.1037/0097-7403.7.3.217 753 Griesbach, G. S., Hu, D., & Amsel, A. (1998). Effects of MK-801 on vicarious trial-and-error and reversal 754 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint of olfactory discrimination learning in weanling rats. Behavioural Brain Research, 97(1–2), 29–38. 755 https://doi.org/10.1016/s0166-4328(98)00015-1 756 Goodlett, C. R., Thomas, J. D., & West, J. R. (1991). Long-term deficits in cerebellar growth and rotarod 757 performance of rats following “binge-like” alcohol exposure during the neonatal brain growth 758 spurt. Neurotoxicology and Teratology, 13(1), 69–74. https://doi.org/10.1016/0892-759 0362(91)90029-V 760 Gursky, Z. H., Savage, L. M., & Klintsova, A. Y. (2019). Nucleus reuniens of the midline thalamus of a rat 761 is specifically damaged after early postnatal alcohol exposure. NeuroReport, 30(10). 762 https://journals.lww.com/neuroreport/Fulltext/2019/07010/Nucleus_reuniens_of_the_midline_thala763 mus_of_a_rat.10.aspx 764 Gursky, Z. H., Savage, L. M., & Klintsova, A. Y. (2021). Executive functioning-specific behavioral 765 impairments in a rat model of human third trimester binge drinking implicate prefrontal-thalamo-766 hippocampal circuitry in Fetal Alcohol Spectrum Disorders. Behavioural Brain Research, 405, 767 113208. https://doi.org/10.1016/j.bbr.2021.113208 768 Gursky, Z. H., Spillman, E. C., & Klintsova, A. Y. (2020). Single-day Postnatal Alcohol Exposure Induces 769 Apoptotic Cell Death and Causes long-term Neuron Loss in Rodent Thalamic Nucleus Reuniens. 770 Neuroscience, 435, 124–134. https://doi.org/10.1016/j.neuroscience.2020.03.046 771 Hallock, H. L., Arreola, A. C., Shaw, C. L., & Griffin, A. L. (2013). Dissociable roles of the dorsal striatum 772 and dorsal hippocampus in conditional discrimination and spatial alternation T-maze tasks. 773 Neurobiology of Learning and Memory, 100, 108–116. https://doi.org/10.1016/j.nlm.2012.12.009 774 Hallock, H. L., Wang, A., & Griffin, A. L. (2016). Ventral Midline Thalamus Is Critical for Hippocampal–775 Prefrontal Synchrony and Spatial Working Memory. The Journal of Neuroscience, 36(32), 8372. 776 https://doi.org/10.1523/JNEUROSCI.0991-16.2016 777 Hamilton, G. F., Hernandez, I. J., Krebs, C. P., Bucko, P. J., & Rhodes, J. S. (2017). Neonatal alcohol 778 exposure reduces number of parvalbumin-positive interneurons in the medial prefrontal cortex 779 and impairs passive avoidance acquisition in mice deficits not rescued from exercise. 780 Neuroscience, 352, 52–63. https://doi.org/10.1016/j.neuroscience.2017.03.058 781 Hamilton, G. F., Whitcher, L. T., & Klintsova, A. Y. (2010). Postnatal binge-like alcohol exposure 782 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint decreases dendritic complexity while increasing the density of mature spines in mPFC Layer II/III 783 pyramidal neurons. Synapse, 64(2), 127–135. https://doi.org/10.1002/syn.20711 784 Hamre, K. M., & West, J. R. (1993). The Effects of the Timing of Ethanol Exposure during the Brain 785 Growth Spurt on the Number of Cerebellar Purkinje and Granule Cell Nuclear Profiles. 786 Alcoholism: Clinical and Experimental Research, 17(3), 610–622. https://doi.org/10.1111/j.1530-787 0277.1993.tb00808.x 788 Hoyme, H. E., Kalberg, W. O., Elliott, A. J., Blankenship, J., Buckley, D., Marais, A.-S., Manning, M. A., 789 Robinson, L. K., Adam, M. P., Abdul-Rahman, O., Jewett, T., Coles, C. D., Chambers, C., Jones, 790 K. L., Adnams, C. M., Shah, P. E., Riley, E. P., Charness, M. E., Warren, K. R., & May, P. A. 791 (2016). Updated Clinical Guidelines for Diagnosing Fetal Alcohol Spectrum Disorders. Pediatrics, 792 138(2), e20154256. https://doi.org/10.1542/peds.2015-4256 793 Hu, D., & Amsel, A. (1995). A simple test of the vicarious trial-and-error hypothesis of hippocampal 794 function. Proceedings of the National Academy of Sciences, 92(12), 5506–5509. 795 https://doi.org/10.1073/pnas.92.12.5506 796 Ikonomidou, C., Bittigau, P., Ishimaru, M. J., Wozniak, D. F., Koch, C., Genz, K., Price, M. T., Stefovska, 797 V., Hörster, F., Tenkova, T., Dikranian, K., & Olney, J. W. (2000). Ethanol-Induced Apoptotic 798 Neurodegeneration and Fetal Alcohol Syndrome. Science, 287(5455), 1056–1060. 799 https://doi.org/10.1126/science.287.5455.1056 800 Ito, H. T., Zhang, S.-J., Witter, M. P., Moser, E. I., & Moser, M.-B. (2015). A prefrontal–thalamo–801 hippocampal circuit for goal-directed spatial navigation. Nature, 522(7554), 50–55. 802 https://doi.org/10.1038/nature14396 803 Janabi-Sharifi, F., Hayward, V., & Chen, C.-S. J. (2000). Discrete-time adaptive windowing for velocity 804 estimation. IEEE Transactions on Control Systems Technology, 8(6), 1003–1009. 805 https://doi.org/10.1109/87.880606 806 Jayachandran, M., Viena, T. D., Garcia, A., Veliz, A. V., Leyva, S., Roldan, V., Vertes, R. P., & Allen, T. A. 807 (2023). Nucleus reuniens transiently synchronizes memory networks at beta frequencies. Nature 808 Communications, 14(1), 4326. https://doi.org/10.1038/s41467-023-40044-z 809 Johnson, A., & Redish, A. D. (2007). Neural Ensembles in CA3 Transiently Encode Paths Forward of the 810 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Animal at a Decision Point. The Journal of Neuroscience, 27(45), 12176. 811 https://doi.org/10.1523/JNEUROSCI.3761-07.2007 812 Jones, M. W., & Wilson, M. A. (2005). Theta Rhythms Coordinate Hippocampal–Prefrontal Interactions in 813 a Spatial Memory Task. PLOS Biology, 3(12), e402. https://doi.org/10.1371/journal.pbio.0030402 814 Kay, K., Chung, J. E., Sosa, M., Schor, J. S., Karlsson, M. P., Larkin, M. C., Liu, D. F., & Frank, L. M. 815 (2020). Constant Sub-second Cycling between Representations of Possible Futures in the 816 Hippocampus. Cell, 180(3), 552-567.e25. https://doi.org/10.1016/j.cell.2020.01.014 817 Kidder, K. S., Miles, J. T., Baker, P. M., Hones, V. I., Gire, D. H., & Mizumori, S. J. Y. (2021). A selective 818 role for the mPFC during choice and deliberation, but not spatial memory retention over short 819 delays. Hippocampus, 31(7), 690–700. https://doi.org/10.1002/hipo.23306 820 Klintsova, A. Y., Cowell, R. M., Swain, R. A., Napper, R. M. A., Goodlett, C. R., & Greenough, W. T. 821 (1998). Therapeutic effects of complex motor training on motor performance deficits induced by 822 neonatal binge-like alcohol exposure in rats: I. Behavioral results. Brain Research, 800(1), 48–61. 823 https://doi.org/10.1016/S0006-8993(98)00495-8 824 Lawrence, R. C., Otero, N. K. H., & Kelly, S. J. (2012). Selective effects of perinatal ethanol exposure in 825 medial prefrontal cortex and nucleus accumbens. Neurotoxicology and Teratology, 34(1), 128–826 135. https://doi.org/10.1016/j.ntt.2011.08.002 827 Layfield, D. M., Patel, M., Hallock, H., & Griffin, A. L. (2015). Inactivation of the nucleus 828 reuniens/rhomboid causes a delay-dependent impairment of spatial working memory. 829 Neurobiology of Learning and Memory, 125, 163–167. https://doi.org/10.1016/j.nlm.2015.09.007 830 Livy, D. J., Miller, E. K., Maier, S. E., & West, J. R. (2003). Fetal alcohol exposure and temporal 831 vulnerability: Effects of binge-like alcohol exposure on the developing rat hippocampus. 832 Neurotoxicology and Teratology, 25(4), 447–458. https://doi.org/10.1016/s0892-0362(03)00030-8 833 Maharjan, D. M., Dai, Y. Y., Glantz, E. H., & Jadhav, S. P. (2018). Disruption of dorsal hippocampal – 834 prefrontal interactions using chemogenetic inactivation impairs spatial learning. Neurobiology of 835 Learning and Memory, 155, 351–360. https://doi.org/10.1016/j.nlm.2018.08.023 836 Mattson, S. N., Bernes, G. A., & Doyle, L. R. (2019). Fetal Alcohol Spectrum Disorders: A Review of the 837 Neurobehavioral Deficits Associated With Prenatal Alcohol Exposure. Alcoholism: Clinical and 838 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Experimental Research, 43(6), 1046–1062. https://doi.org/10.1111/acer.14040 839 Miles, J. T., Mullins, G. L., & Mizumori, S. J. Y. (2024). Flexible decision-making is related to strategy 840 learning, vicarious trial and error, and medial prefrontal rhythms during spatial set-shifting. 841 Learning & Memory (Cold Spring Harbor, N.Y.), 31(7), a053911. 842 https://doi.org/10.1101/lm.053911.123 843 Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortex function. Annual Review of 844 Neuroscience, 24, 167–202. https://doi.org/10.1146/annurev.neuro.24.1.167 845 Muenzinger, K. F. (1938). Vicarious Trial and Error at a Point of Choice: I. A General Survey of its 846 Relation to Learning Efficiency. The Pedagogical Seminary and Journal of Genetic Psychology, 847 53(1), 75–86. https://doi.org/10.1080/08856559.1938.10533799 848 Murawski, N. J., Klintsova, A. Y., & Stanton, M. E. (2012). Neonatal alcohol exposure and the 849 hippocampus in developing male rats: Effects on behaviorally induced CA1 c-Fos expression, 850 CA1 pyramidal cell number, and contextual fear conditioning. Neuroscience, 206, 89–99. 851 https://doi.org/10.1016/j.neuroscience.2012.01.006 852 O’Neill, P.-K., Gordon, J. A., & Sigurdsson, T. (2013). Theta oscillations in the medial prefrontal cortex are 853 modulated by spatial working memory and synchronize with the hippocampus through its ventral 854 subregion. The Journal of Neuroscience, 33(35), 14211–14224. 855 https://doi.org/10.1523/JNEUROSCI.2378-13.2013 856 Otero, N. K. H., Thomas, J. D., Saski, C. A., Xia, X., & Kelly, S. J. (2012). Choline Supplementation and 857 DNA Methylation in the Hippocampus and Prefrontal Cortex of Rats Exposed to Alcohol During 858 Development. Alcohol: Clinical and Experimental Research, 36(10), 1701–1709. 859 https://doi.org/10.1111/j.1530-0277.2012.01784.x 860 Papale, A. E., Stott, J. J., Powell, N. J., Regier, P. S., & Redish, A. D. (2012). Interactions between 861 deliberation and delay-discounting in rats. Cognitive, Affective, & Behavioral Neuroscience, 12(3), 862 513–526. https://doi.org/10.3758/s13415-012-0097-7 863 Popova, S., Charness, M. E., Burd, L., Crawford, A., Hoyme, H. E., Mukherjee, R. A. S., Riley, E. P., & 864 Elliott, E. J. (2023). Fetal alcohol spectrum disorders. Nature Reviews Disease Primers, 9(1), 11. 865 https://doi.org/10.1038/s41572-023-00420-x 866 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Rasmussen, C. (2005). Executive Functioning and Working Memory in Fetal Alcohol Spectrum Disorder. 867 Alcoholism: Clinical and Experimental Research, 29(8), 1359–1367. 868 https://doi.org/10.1097/01.alc.0000175040.91007.d0 869 Redish, A. D. (2016). Vicarious trial and error. Nature Reviews Neuroscience, 17(3), 147–159. 870 https://doi.org/10.1038/nrn.2015.30 871 Sangiamo, D. T., Warren, M. R., & Neunuebel, J. P. (2020). Ultrasonic signals associated with different 872 types of social behavior of mice. Nature Neuroscience, 23(3), 411–422. 873 https://doi.org/10.1038/s41593-020-0584-z 874 Schmidt, B., Duin, A. A., & Redish, A. D. (2019). Disrupting the medial prefrontal cortex alters 875 hippocampal sequences during deliberative decision making. Journal of Neurophysiology, 121(6), 876 1981–2000. https://doi.org/10.1152/jn.00793.2018 877 Schmidt, B., Papale, A., Redish, A. D., & Markus, E. J. (2013). Conflict between place and response 878 navigation strategies: Effects on vicarious trial and error (VTE) behaviors. Learning & Memory, 879 20(3), 130–138. https://doi.org/10.1101/lm.028753.112 880 Steiner, A. P., & Redish, A. D. (2012). The Road Not Taken: Neural Correlates of Decision Making in 881 Orbitofrontal Cortex. Frontiers in Neuroscience, 6. 882 https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2012.00131 883 Stout J. J., George A. E., Kim S., Hallock H. L., Griffin A. L. (2023). Using synchronized brain rhythms to 884 bias memory-guided decisions. eLife 12:RP92033. https://doi.org/10.7554/eLife.92033.2 885 Stout, J. J., Hallock, H. L., George, A. E., Adiraju, S. S., & Griffin, A. L. (2022). The ventral midline 886 thalamus coordinates prefrontal–hippocampal neural synchrony during vicarious trial and error. 887 Scientific Reports, 12(1), 10940. https://doi.org/10.1038/s41598-022-14707-8 888 Tang, W., Shin, J. D., & Jadhav, S. P. (2021). Multiple time-scales of decision-making in the hippocampus 889 and prefrontal cortex. eLife, 10, e66227. https://doi.org/10.7554/eLife.66227 890 Thomas, J. D., Wasserman, E. A., West, J. R., & Goodlett, C. R. (1996). Behavioral deficits induced by 891 bingelike exposure to alcohol in neonatal rats: Importance of developmental timing and number of 892 episodes. Developmental Psychobiology, 29(5), 433–452. https://doi.org/10.1002/(SICI)1098-893 2302(199607)29:53.0.CO;2-P 894 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Thomas, J. D., Weinert, S. P., Sharif, S., & Riley, E. P. (1997). MK-801 Administration During Ethanol 895 Withdrawal in Neonatal Rat Pups Attenuates Ethanol-Induced Behavioral Deficits. Alcohol: 896 Clinical and Experimental Research, 21(7), 1218–1225. https://doi.org/10.1111/j.1530-897 0277.1997.tb04441.x 898 Tolman, E. C. (1939). Prediction of vicarious trial and error by means of the schematic sowbug. 899 Psychological Review, 46(4), 318–336. 900 Tran, T. D., & Kelly, S. J. (2003). Critical periods for ethanol-induced cell loss in the hippocampal 901 formation. Neurotoxicology and Teratology, 25(5), 519–528. https://doi.org/10.1016/S0892-902 0362(03)00074-6 903 Viena, T. D., Linley, S. B., & Vertes, R. P. (2018). Inactivation of nucleus reuniens impairs spatial working 904 memory and behavioral flexibility in the rat. Hippocampus, 28(4), 297–311. 905 https://doi.org/10.1002/hipo.22831 906 Wang, G.-W., & Cai, J.-X. (2006). Disconnection of the hippocampal–prefrontal cortical circuits impairs 907 spatial working memory performance in rats. Behavioural Brain Research, 175(2), 329–336. 908 https://doi.org/10.1016/j.bbr.2006.09.002 909 Wang, J. X., Cohen, N. J., & Voss, J. L. (2015). Covert rapid action-memory simulation (CRAMS): A 910 hypothesis of hippocampal–prefrontal interactions for adaptive behavior. Memory and Decision 911 Making, 117, 22–33. https://doi.org/10.1016/j.nlm.2014.04.003 912 Whitcher, L. T., & Klintsova, A. Y. (2008). Postnatal binge-like alcohol exposure reduces spine density 913 without affecting dendritic morphology in rat mPFC. Synapse, 62(8), 566–573. 914 https://doi.org/10.1002/syn.20532 915 Wozniak, D. F., Hartman, R. E., Boyle, M. P., Vogt, S. K., Brooks, A. R., Tenkova, T., Young, C., Olney, J. 916 W., & Muglia, L. J. (2004). Apoptotic neurodegeneration induced by ethanol in neonatal mice is 917 associated with profound learning/memory deficits in juveniles followed by progressive functional 918 recovery in adults. Neurobiology of Disease, 17(3), 403–414. 919 https://doi.org/10.1016/j.nbd.2004.08.006 920 921 922 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Figures 923 924 Figure 1. The alcohol exposed group spends less time in the choice point than the sham intubated 925 group. A) Experimental timeline. PD=postnatal day, AE=alcohol exposed, SI=sham intubated, 926 CA=continuous alternation, DA=delayed alternation. B) CA task schematic. Rats alternated between left 927 and right choices over trials to receive a reward. C) DA task schematic. After each trial, rats returned to 928 the start box (gray circle) to complete a delay of either 10, 30, or 60 seconds (s). D) DA task choice 929 accuracy for all 10-, 30-, and 60-second delay trials in the AE (red) and SI (blue) groups. The proportion 930 of correct trials decreases with delay length in the SI and AE groups and is not different between groups. 931 E) Rats in the AE group (red) spend significantly less time in the choice point compared to rats in the SI 932 group (blue) during the DA task. Colored dots indicate individual rats. An outlier rat in the AE group is 933 indicated with a red “X”. Inset: T-maze with the choice point highlighted in pink. *p<0.05, **p<0.01, 934 ***p<0.001. Error bars represent mean +/- standard error of the mean. 935 936 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint 937 Figure 2. Vicarious trial and error behaviors are less frequent in the alcohol exposed group 938 compared to the sham intubated group. A) DA task zlnIdPhi distribution based on choice point tracking 939 data. The VTE threshold (zlnIdPhi= 0.3; red dashed line) was determined as the point where the zlnIdPhi 940 distribution deviated from a normal distribution. Inset: Example non-VTE trial. Trial trajectory overlays 941 tracking data from an example recording session (light gray). Trajectory color represents the normalized 942 velocity of the rat. B) Left: Method 1 of VTE trial visualization. Example VTE trial with zlnIdPhi score 943 above threshold. Right: Method 2 of VTE trial visualization. Example VTE trial with zlnIdPhi score below 944 threshold but where the rat enters both goal arms (black boxes). This method allowed us to identify VTE 945 trials that would have originally been excluded due to high velocity through the choice point. Both Method 946 1 and Method 2 were used to identify VTE trials (see Methods for details). C) The overall proportion of 947 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint trials with a VTE in the choice point is lower in the AE (red) group across delays compared to the SI (blue) 948 group. D) Left: The AE group shows fewer VTEs in the stem of the T-maze than the SI group. The 949 proportion of VTE trials is not affected by delay length. Right) Example trial with VTEs (indicated with 950 arrows) in the T-maze stem. E) Choice accuracy on the subset of trials with VTEs at the choice point. 951 Compared to choice accuracy on all trials, accuracy on VTE trials did not decrease with increased delay 952 length. F) Choice accuracy on non-VTE trials, trials with a VTE in the stem (stem VTE) and trials with a 953 VTE at the choice point (choice point VTE) collapsed across delay. While there was no significant 954 difference in accuracy between the AE and SI groups, there was a main effect of trial type on choice 955 accuracy such that accuracy was significantly lower on choice point VTE trials compared to stem VTE 956 and non-VTE trials. G) AE and SI groups show similar proportions of VTE trials at the choice point during 957 the CA task. Colored dots indicate individual rats. *p<0.05, ***p<0.001. Error bars represent mean +/- 958 standard error of the mean. 959 960 961 Figure 3. The proportion of error trials with vicarious trial and errors decreases with delay 962 duration and is lower in the alcohol exposed group compared to the sham intubated group. A) 963 Schematic of VTE Error (top) and VTE Correct (bottom) trials. Choice point trajectories from example 964 trials are represented in black. L=left choice, R=right choice. The choice point is highlighted in pink. B) 965 The proportion of VTE error trials is lower in the AE group (red) compared to the SI group (blue). VTE 966 error trials decrease with delay duration, with the highest proportion of VTEs occurring on 10-second 967 delay error trials. C) The proportion of VTE correct trials is not significantly different between groups. VTE 968 trial proportion is not affected by delay on correct trials. B-C) VTEs occur on a greater proportion of error 969 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint trials compared to correct trials. ***p<0.001. Data are represented as mean +/- standard error of the 970 mean. 971 972 973 Figure 4. The frequency of vicarious trial and errors decreases with experience in the sham 974 intubated group, but not the alcohol exposed group. A) Scatterplot demonstrating a significant 975 negative correlation between the proportion of VTEs and session choice accuracy in the AE (red) and SI 976 (blue) groups. To directly compare the relationship between VTEs and session accuracy, only trials with 977 position data (and therefore VTE data) were included in the calculation of a session choice accuracy 978 average. B) While VTE proportion decreases over sessions in the SI group, the AE group does not show 979 a change in VTE frequency. C) Choice accuracy is not significantly different between groups across 980 sessions on the DA task. Both the SI and AE groups show improvements across sessions, shown as 981 significant positive correlations between session number and choice accuracy. Data are represented as 982 mean +/- standard error of the mean. *p<0.05, **p<0.01, ***p<0.001. 983 984 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint 985 Figure 5. Vicarious trial and errors are associated with perseverative errors in the alcohol exposed 986 group. A) The AE group (red) performs poorer on the trial following 10-second delay trials which included 987 VTEs compared to the SI group (blue). In contrast, both groups perform similarly following 30-second (B) 988 and 60-second (C) delay VTE trials. Colored dots indicate individual rats. D) AE and SI groups perform 989 similarly on the trial following 10-second delay trials that were non-VTE trials. E) Delay distributions of the 990 trial following a VTE during 10-second, 30-second, and 60-second delay trials. Due to the delay sequence 991 followed during the DA task, 10-second delays and 60-second delays were never followed by consecutive 992 delays of the same length. Delay distributions of the SI group are boxed in blue and delay distributions of 993 the AE group are boxed in red. 10-second delay trials are indicated in purple, 30-second delay trials are 994 indicated in pink, and 60-second delay trials are indicated in green. F) Rats in the SI and AE groups 995 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint perform similar proportions of perseverative error trials during the DA task. G) The proportions of VTEs 996 and perseverative errors are positively correlated at the rat (left) and session (right) levels in the AE group 997 (red) but not the SI group (blue). *p<0.05. ***p<0.001. Bar plots represent the mean +/- standard error of 998 the mean. 999 1000 1001 Figure 6. mPFC theta power and HPC beta power are altered after alcohol exposure. A) LFPs were 1002 recorded from the mPFC and HPC during choice point occupancy (highlighted in pink) on VTE trials. 1003 Example signals from the mPFC (green) and HPC (purple) are shown to the right. B) Left: mPFC power 1004 distribution as a function of frequency for the AE (red) and SI (blue) groups. The mean power distribution 1005 is represented as a solid line and the standard error of the mean is represented as the shaded area 1006 around the mean. Analyses were performed over the 6-10 Hz theta range (highlighted in yellow) and the 1007 15-30 Hz beta range (highlighted in pink). Middle: Bar plot demonstrating mPFC theta power during VTEs 1008 is lower in the AE group compared to the SI group. Right: Bar plot showing mPFC beta power during 1009 VTEs is not different between groups. Bar plots represent the mean +/- standard error of the mean. 1010 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint Colored dots indicate individual rats. C) Same as B, except for HPC power. HPC theta power is not 1011 different between groups (middle), while beta power is higher in the AE group compared to the SI group 1012 (right). D) Same as B, except for mPFC-HPC coherence. mPFC-HPC theta (middle) and beta (right) 1013 coherence are not different between groups. E) Left: mPFC theta power is lower in the AE group 1014 compared to the SI group during non-VTE trials. Two outlier rats were identified in the AE group and are 1015 indicated with a red “X”. Right: mPFC beta power is not different between groups during non-VTE trials. 1016 F) HPC theta power is not significantly different between groups during non-VTE trials (left), whereas 1017 HPC beta power is higher in the AE group than the SI group (right). G) mPFC-HPC theta (left) and beta 1018 (right) coherence are not different between groups during non-VTE trials. H) Scatterplot showing that the 1019 proportion of VTE trials is negatively correlated with mPFC theta power during VTE trials in the AE group 1020 but not the SI group. Only trials with clean LFP data were considered in the calculation of VTE trial 1021 proportion. *p<0.05. 1022 1023 1024 Figure 7. The prevalence of mPFC-HPC synchronous events is altered after alcohol exposure. A) 1025 Left: Schematic of moving window method to calculate mPFC-HPC coherence. Coherence was 1026 calculated over 1.25-second (s) events that were gradually shifted by 250 milliseconds (ms). Example 1027 LFPs from the mPFC and HPC are represented in green and purple, respectively. Right: Stem plots 1028 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint showing theta coherence across events from example trials of different AE (red) and SI (blue) rats. Each 1029 trial has a magnitude coherence of 0.4 during choice point occupancy. Note that the degree of mPFC-1030 HPC synchronization varies across events within this period. B) Bar plot demonstrating that magnitude 1031 coherence (6-10 Hz) is not different between the AE (red) and SI (blue) groups when calculated with the 1032 moving window approach. Colored dots indicate individual rats. C) CDF plot showing that the distributions 1033 of mPFC-HPC theta coherence events (6-10 Hz) are significantly different between the AE (red) and SI 1034 (blue) groups during VTEs. D) Same as C, except coherence events were measured from non-VTE trials. 1035 E) CDF plot showing theta coherence event distributions of individual AE rats compared to the theta 1036 coherence event distribution of the SI group (dark blue line). Asterisks in the legend indicate that the 1037 coherence event distribution of the corresponding rat is significantly different from the coherence event 1038 distribution of the SI group. *p<0.05, **p<0.01, ***p<0.001. 1039 1040 1041 Figure 8. Using supervised machine learning to predict treatment with behavioral data. A) 1042 Determining K for KNN Classifier. As accuracy initially plateaus at 7 (blue line), this value was chosen as 1043 K. B) The KNN Classifier performs at 76.5% accuracy (purple line) when using the true labels to predict 1044 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint the treatment of AE and SI rats. Accuracy is significantly above chance levels as determined by 1045 performing a z-test with the accuracy distribution obtained by shuffling the labels of AE and SI rats (mean 1046 accuracy represented with a black dashed line). C) Both a KNN Classifier (purple) and a Euclidean 1047 Classifier (pink) achieve the same accuracy when predicting the treatment of AE and SI rats. D) Iteratively 1048 removing categories from the KNN Classifier to determine which categories contribute to accurate 1049 classification. Time spent in the choice point, the proportion of VTE error trials by delay, and overall 1050 choice accuracy on the DA task by delay are the only categories that when removed result in reduced 1051 classifier accuracy (not significantly different from chance levels). Dashed lines represent mean accuracy 1052 of the shuffled distributions. E) Using only time spent in the choice point, the proportion of VTE error trials, 1053 and overall choice accuracy, the classifier performs at identical accuracy as when all categories are 1054 included (C) and performs significantly above chance levels. Testing the classifier on all remaining 1055 categories results in accuracy that is not significantly different from chance levels. pVTE=proportion of 1056 VTE trials. pCorrect= proportion of correct trials. *p<0.05 compared to shuffled distribution. 1057 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 29, 2024. ; https://doi.org/10.1101/2024.07.28.605480doi: bioRxiv preprint