{"paper_id":"01eda6aa-b683-48d1-b69d-a02227b50d98","body_text":"1 \nNeural Representation of Associative Threat Learning in Pulvinar 1 \nDivisions, Lateral Geniculate Nucleus, and Mediodorsal Thalamus in 2 \nHumans 3 \n 4 \n 5 \nMuhammad Badarnee1, Zhenfu Wen1, B. Isabel Moallem1, Stephen 6 \nMaren2,3, and Mohammed R Milad1* 7 \n 8 \n 9 \n1Department of Psychiatry and Behavioral Sciences, The University of 10 \nTexas, Health Science Center at Houston, McGovern Medical School, TX 11 \nUSA. 12 \n2Beckman Institute for Advanced Science and Technology, University of 13 \nIllinois Urbana-Champaign, Urbana, IL 14 \n3Department of Psychology, University of Illinois Urbana-Champaign, 15 \nChampaign, IL 16 \n 17 \n 18 \n 19 \n*Corresponding Author: 20 \nMohammed R Milad, PhD 21 \nDepartment of Psychiatry and Behavioral Sciences 22 \nUTHealth, McGovern Medical School  23 \n1941 East Road, BBSB - 2118  24 \nHouston, Texas 77054 25 \nmohammed.r.milad@uth.tmc.edu 26 \n 27 \n 28 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 2 \nAbstract  29 \nUnderstanding the neural mechanisms underlying associative threat learning is essential 30 \nfor advancing behavioral models  of threat and adaptation. We investigated distinct 31 \nactivation patterns across thalamic pulvinar divisions, lateral geniculate nucleus (LGN), 32 \nand mediodorsal thalamus (MD) during the acquisition of associative threat learning in 33 \nthe MRI. We revealed parallel thalamic learning systems within the anterior pulvinar and 34 \nMD, supporting distinct mechanisms of automatic survival vs. more deliberate learning. 35 \nAdditionally, o ur findings support a novel  hierarchical pulvinar model  during fear 36 \nconditioning: the medial pulvinar mediate s basic threat information from the inferior and 37 \nlateral divisions to the anterior pulvinar for integrative learning. Pulvinar divisions and MD 38 \nsupport extinction learning . These regions also process salience and modulat e 39 \nsafe/threat memory expression during extinction recall and threat renewal . The LGN 40 \nsustains feedforward processing of anticipated visual input throughout all threat phases. 41 \nThis study extends dominant brain models of threat learning and memory, reframing our 42 \nunderstanding of distinct thalamic roles in these psychological processes.  43 \nKeywords: Thalamus; Pulvinar; Lateral geniculate nucleus; Mediodorsal thalamus; 44 \nThreat learning; Conditioning; Extinction; Recall extinction; Threat renewal; Emotional 45 \nmemory.  46 \n 47 \n 48 \n 49 \n 50 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 3 \nIntroduction   51 \n 52 \n 53 \nOver a century ago, Ivan Pavlov provided foundational behavioral evidence for 54 \nassociative learning, demonstrating that pairing a neutral stimulus with an unconditioned 55 \nstimulus could elicit a conditioned response(Pavlov, 1904 ). This  discovery laid the 56 \ngroundwork for modern learning theories and has been particularly influential in 57 \nunderstanding how humans learn to associate neutral stimuli with threat s, a process 58 \nknown as associative threat learning . This survival-oriented learning plays a key role in 59 \nshaping a broad range of behavioral and emotional response s, such as avoidance, 60 \ndecision-making under threats, and fear regulation(Badarnee et al., 2025; Kolling et al., 61 \n2014; Korn and Bach, 2019, 2018; Milad et al., 2013) . Research on t he neural 62 \nmechanisms underlying associative threat learning has primarily focused on brain regions 63 \ninvolved in emotional regulation, such as the amygdala, hippocampus, and prefrontal 64 \ncortex (PFC)(Fullana et al., 2015; Krasne et al., 2021; Milad and Quirk, 2012; Vuilleumier 65 \net al., 2003) . More r ecently, the thalamus has received growing attention in threat 66 \nlearning(Lithari et al., 2015; Penzo et al., 2015; Ramanathan et al., 2018; Ramanathan 67 \nand Maren, 2019; Ratigan et al., 2023; Totty et al., 2023) , since its role has been 68 \nreconsidered beyond the traditional view of a mere relay station. Yet, the distinct 69 \ncontribution of this complex structure to associative threat learning remains poorly 70 \nunderstood.  71 \n 72 \nThe thalamus is a hub structure in the mammalian brain that plays multiple critical roles, 73 \nranging from basic sensory processing  to higher-order functions and threat 74 \nlearning(Halassa and Sherman, 2019; Hummos  et al., 2022; Hwang et al., 2017; 75 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 4 \nSaalmann and Kastner, 2011; Sherman, 2016, 2007) . It has long  been proposed that 76 \nthreat detection is mediated by the pulvinar and the lateral geniculate nucleus (LGN)(Carr, 77 \n2015; Pessoa and Adolphs, 2010; Silverstein and Ingvar, 2015) , two major nuclei of the 78 \nvisual thalamus (Arcaro et al., 2015; Casanova and Chalupa, 2023; Takakuwa et al., 79 \n2021). The pulvinar projects directly to the amygdala and is believed to subconsciously 80 \nfacilitate rapid survival reactions to threats via a subcortical path (‘low road’)(Carr, 2015; 81 \nPessoa and Adolphs, 2010; Rafal et al., 2015; Wei et al., 2015) . The LGN, on the other 82 \nhand, is part of a slower but more accurate neural pathway , connecting peripheral 83 \ninformation to cortical regions for comprehensive processing (‘high road’). Additionally, 84 \nthe involvement of the mediodorsal thalamus (MD) in threat learning has also been 85 \nproposed(Lee and Shin, 2016; Lee et al., 2011) . Although this nucleus is not typically 86 \nclassified as part of the visual thalamus , it serves as a high-order region reciprocally 87 \nconnected to the amygdala and PFC  while playing a critical role in executive 88 \nfunctions(Hwang et al., 2020; Li et al., 2022; Mukherjee et al., 2021; Wolff and Halassa, 89 \n2024).  90 \n 91 \nThe general implication of these nuclei in threat processing has been demonstrated in 92 \nprimates and rodent models. It has been shown that exposing primates to snake images 93 \nelicited increased neuronal firing within the pulvinar(Le et al., 2013). Excitation of neurons 94 \nwithin the LGN enhanced the acquisition of eyeblink conditioning in rodents (Halverson 95 \nand Freeman, 2010) . Inhibiting this nucleus was associated with impaired conditioned 96 \nresponses(Shi and Davis, 2001; Steinmetz et al., 2013), and activating GABA neurons in 97 \nthe LGN reduced freezing response to an overhead dark shadow that mimics a real-world 98 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 5 \npredator in rats(Salay and Huberman, 2021). Reduced freezing was also observed after 99 \nMD lesions(Li et al., 2004)  and when the connections with the anterior cingulate cortex 100 \nwere ablated(Zheng et al., 2020). Lesions within the MD are also associated with impaired 101 \nfear extinction(Lee et al., 2011) , and injecting gabazine , a modulator of extra-synaptic 102 \nGABA receptors, into the MD facilitated extinction(Paydar et al., 2014). 103 \n 104 \nEvidence regarding the involvement of the thalamus in the context of threat learning in 105 \nhumans is relatively sparse. Much of our knowledge comes from studies focusing on 106 \nattention, perception, and decision -making. For example, the LGN has been mainly 107 \nviewed as a transmitter of peripheral information to other brain regions and found to be 108 \nassociated with selective attention and anticipation of visual stimuli (Mahoney and 109 \nSchmidt, 2024; O’Connor et al., 2002; Saalmann and Kastner, 2009). The MD contribution 110 \nto perception, spatial attention, and decision -making has also been reported (Griffiths et 111 \nal., 2022; Wurtz et al., 2011) . Beyond cognitive research, s ome human studies have 112 \nspecifically investigated the pulvinar ’s engagement in threat detection. Individuals with 113 \nincreased fiber density in the pulvinar-amygdala pathway showed an enhanced ability to 114 \nrecognize fearful faces(McFadyen et al., 2019). A patient with a complete lesion in the left 115 \npulvinar showed slower responses to threatening images when stimuli were presented 116 \non the ipsilesional, but not contralesional, field(Ward et al., 2005). In agreement with this, 117 \nunseen stimuli (fearful vs. happy faces) presented on the blind side of patients with 118 \nhemianopia moderated their performance when the pulvinar was spared but not when it 119 \nwas lesioned(Bertini et al., 2018).  120 \n 121 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 6 \nSpecific thalamic contribution to  threat processing is still largely unknown, and 122 \ntranslational research on this topic is limited. We aim to address these critical gaps in our 123 \nunderstanding by investigating the distinct neural representations of associative threat 124 \nlearning in the human pulvinar divisions, LGN, and MD. We analyzed the neuroimaging 125 \ndata of 293 -412 controls. All participants underwent a two -day threat learning 126 \nparadigm(Milad et al., 2009, 2007; Wen et al., 2024, 2022) while in the fMRI scanner. The 127 \nconditioned stimulus ( CS+) (e.g., red light) was associated with an electric shock 128 \n(unconditioned stimulus US, 62.5% reinforcement rate), while a control light (e.g., blue) 129 \nwas never paired with the US (CS -). This conditioning phase occurred in a computer-130 \ndisplayed visual context A (e.g., an office).  The extinction learning included presenting 131 \nthe CS+ and CS - with no US reinforcement in a distinct context B (e.g., bookcase). In 132 \nextinction recall and threat renewal, the extinguished stimuli were presented with no US 133 \nreinforcement within contexts B (safe contextual cues) and A ( threat contextual cues), 134 \nrespectively. Schematic illustrations of each phase of the paradigm are presented in Figs. 135 \n1a, 4a-6a.  136 \n 137 \nWe focus on neural activation patterns within pulvinar divisions, LGN, and MD while 138 \nacquiring the CS-US association. As associative learning is a rapid psychological 139 \nprocess(Konrad et al., 2024), we analyzed the first four trials of all threat learning phases. 140 \nWe compared brain activation (BOLD) to the CS+ vs. CS- at the block level by averaging 141 \nthe activation across all four trials and at a trial-by-trial level to provide finer activation 142 \ntemporal resolution. This dual approach allows us to capture the general neural 143 \nresponses associated with associative threat learning and provides new insights into trial-144 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 7 \nlevel dynamics (Wen et al., 2022) . The current dominant neurocognitive thalamic 145 \nmodels(Sherman, 2007; Sherman and Guillery, 2006)  highlight the thalamus’s role in  146 \nmediating cortical-cortical communications and facilitating high-order functions. Based on 147 \nthis view, we anticipate distinct pulvinar, MD, and LGN roles. The LGN, as a first -order 148 \nnucleus, is  expected to serve relay function s, transmitting information with minimal 149 \nintegrations. The pulvinar and MD, on the other hand, as higher -order nuclei, are likely 150 \ninvolved in more complex processing, potentially integrating threat-related information. 151 \n 152 \nResults  153 \nParallel functional representation of associative threat learning in the anterior 154 \npulvinar and MD during conditioning 155 \nThe anterior pulvinar and MD contribute to early associative threat learning, as evidenced 156 \nby increased functional activation in response to CS+ compared to CS- at the block level 157 \n(Fig. 1b-c). The trial-wise analysis revealed a distinct activation pattern. In the first trial, 158 \nwe observed similar activation levels for both CS+ and CS- (no significant difference), but 159 \nin the subsequent trial, the BOLD signal in both regions was  heightened specifically for 160 \nCS+ (Fig. 1b-c, Extended Tables 1-2). In our paradigm, the electric shock was paired with 161 \nCS+ at the end of the CS presentation. The similarity in BOLD responses to CS+ and CS- 162 \nduring the first trial likely reflects an initial equivalence in the emotional valence of the 163 \nstimuli. The gradual increase in activation for CS+ by the second trial suggests a shift in 164 \nthe emotional valence, indicating rapid associative learning.  165 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 8 \nThe similarity in trial-level activation patterns in the anterior pulvinar and MD raises the 166 \nquestion of whether this apparent similarity truly reflects a parallel functional contribution 167 \nto associative threat learning . To test this possibility, w e first quantified the trial-wise 168 \nrelationships using Pearson correlation coefficients between the two regions . The 169 \nobserved correlations support consistent co -activation across all corresponding trials (r 170 \nvalues ≥ 0.66, p < 0.001, Fig. 2a, left). However, this approach could not exclude the 171 \npossibility that the shared variance is a consequence of shared anatomical proximity , 172 \nparticularly because both regions are part of the same brain structure, i.e., the thalamus. 173 \nTo control for this confound, we conducted a hierarchical regression model in four steps 174 \nfor each trial . We modeled MD activation as the target and progressively ad ded the 175 \npulvinar division s as covariates . Despite small variations in regression weights (beta 176 \nvalues) across control models, the anterior pulvinar consistently exhibited the highest 177 \nbeta estimat es (Fig. 2a, middle ). Importantly, these values remain consistent across 178 \n10,000 bootstrapping iterations with replacements, suggesting that the anterior pulvinar 179 \neffect is robust and stable across resampled datasets  (Fig. 2a, right). Additionally, the 180 \nanterior pulvinar explained the largest proportion of variance  in MD activation  (r2 = 43-181 \n60.7%) relative to other pulvinar divisions, which negligibly contributed to this parameter 182 \n(pie charts, Fig. 2a). Finally, applying the same analytical approach while controlling for 183 \nthe LGN activation as an alternative anatomical control beyond the pulvinar itself, yielded 184 \nsimilar results (Fig. 2b). Together, these analyses support preferential trial-wise functional 185 \nco-activations between the anterior pulvinar and MD , independent of shared anatomical 186 \nproximity.  187 \n 188 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 9 \nAlthough co-activation does not necessarily imply similar activation magnitude, we further 189 \ntested the differences in activation levels in the anterior pulvinar and MD. We used a t-190 \ntest to capture the differences in the overall activation at the block level and repeated 191 \nmeasures analysis of variance ( RM-ANOVA) to capture the activation differences at the 192 \ntrial level. The results showed that the overall MD activation in response to CS+ was 193 \nhigher than the anterior pulvinar response  (p-value < 0.001, Fig. 2c). The trial -wise 194 \nanalysis revealed heightened activation in the MD compared to the anterior pulvinar , 195 \nparticularly during trials 3 and 4 (pFDR-values < 0.05, Fig. 2c). The increased activation 196 \nwithin the MD potentially reflects its role in deeper processing of threat  relatively to the 197 \nanterior pulvinar. 198 \n 199 \nA data -driven approach reveals hierarchical functional processing of threat in 200 \npulvinar divisions 201 \nThe anatomical and functional interconnections between pulvinar divisions remain 202 \ninsufficiently characterized. To address this gap and understand the pulvinar’s role in 203 \nearly associative threat learning, we integrated findings from complementary analyses 204 \ninto a novel model designed to investigate the functional relationships between pulvinar 205 \ndivisions. 206 \n 207 \nWe first defined the activation within each pulvinar division as the differences in BOLD 208 \nsignal at the block level in response to CS+ compared to CS -. We then performed RM -209 \nANOVA to test whether the pulvinar divisions  were engaged with different activation 210 \nlevels. We found no activation differences, suggesting similar processing levels of CS+ 211 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 10 \ninformation across all pulvinar divisions (Fig. 3a). Using network analysis, w e explored 212 \nthe underlying functional dynamics between the divisions. We employed the EBICglasso 213 \nmethod(Foygel and Drton, 2010)  to estimate a sparse Gaussian graphical model.  To 214 \nassess the stability of the network and provide robust estimates of edges and centrality 215 \nmeasures, we conducted 10,000 bootstrap resamples. This approach allowed us to 216 \ncompute 95% confidence intervals, ensuring the reliability of our findings. The combined 217 \nuse of EBICglasso and bootstrapping is essential for accurately capturing the dynamics 218 \nof pulvinar division interactions. The resulting network included four nodes and five edges. 219 \nThe network plot pointed to stable direct edges connecting the lateral pulvinar to the 220 \nanterior pulvinar and indirect edges connecting both the lateral and inferior pulvinar to the 221 \nanterior through the medial pulvinar (Fig. 3b). Additionally, the medial pulvinar exhibited 222 \nthe highest centrality values, a measure that  captures the node importance within a 223 \nnetwork, indicating a potential hub role of this division in associative threat learning (Fig. 224 \n3b).  225 \nThese underlying dynamics and the activation timing in the trial-wise analyses (overview 226 \nin Fig. 3c and details in Fig. 1b) led us to propose a functional model of the relationships 227 \nbetween pulvinar divisions. Specifically, the increased activation induced by CS+ during 228 \ntrial 1 in the medial, inferior, and lateral pulvinar suggests that these divisions process 229 \nCS+ information before the anterior pulvinar, which exhibited a delayed BOLD response 230 \nstarting in trial 2. The medial pulvinar mediates connections within the pulvinar network 231 \nand is associated with elevated centrality values. Together, these findings highlight the 232 \nhub role of the medial pulvinar , integrating CS+ information at a higher level , compared 233 \nto sensory-driven processing in the inferior and lateral pulvinar. This possible hierarchical 234 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 11 \norganization is supported by previous evidence showing that the inferior and lateral 235 \npulvinar are associated with processing basic sensory information(Berman and Wurtz, 236 \n2010, 2008; Cortes et al., 2024) , while the medial pulvinar is implicated in higher -order 237 \nprocessing, including attention and working memory(Homman-Ludiye and Bourne, 2019) 238 \n(Fig. 3d).  239 \nThis data-driven approach provides a new perspective on the pulvinar divisions’ functional 240 \nspecialization during associative threat learning.  We hypothesize that, during this 241 \nprocess, the activation in the medial pulvinar mediates the functional relationships 242 \nbetween the inferior and lateral divisions with the anterior pulvinar (Fig. 3e). To test this, 243 \nwe conducted a mediation model analysis and evaluated the model robustness using the 244 \nk-fold cross-validation method. The sample ( N = 293) was randomly divided into three 245 \ngroups of 91, 96, and 106 subjects. Each sub -sample was used to test the model while 246 \nthe remaining groups served as a training phase. This resulted in six iterations of the 247 \nmediation model. The 95% Confidence Intervals for each path coefficient were estimated 248 \nusing 10,000 bootstrapping replications. This choice was made to achieve greater 249 \nprecision and stability in the confidence interval estimates. Finally, applying the mediation 250 \nanalysis to another independent cohort (N = 114) confirmed the model’s applicability. The 251 \nresults in Figs. 3f-3g support the mediating role of the medial pulvinar, demonstrating 252 \nsignificant indirect paths from the lateral and inferior divisions to the anterior pulvinar. 253 \nDetailed parameters are presented in Extended Table 3.  254 \nPulvinar divisions and MD support extinction learning while preserving threat 255 \nmemory across contextual cues   256 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 12 \nAfter our comprehensive examination of the thalamic contribution to threat conditioning, 257 \nwe moved forward to examine the neural contributions of the same regions to extinction 258 \nlearning. During this phase, w e found higher activation in response to CS+ than CS- 259 \nacross all pulvinar divisions  and MD at the block level. The trial -wise analysis 260 \ndemonstrated that these differences are mainly driven by the first two trials (Figs. 4b and 261 \n4c). Together, the pulvinar divisions and MD showed a similar activation pattern where 262 \nthe BOLD signal induced by the CS+ is significantly diminished after trial 2 , highlighting 263 \nthese nuclei’s engagement  in rapid extinction learning.  Detailed parameters are 264 \npresented in Extended Tables 4-5.  265 \n 266 \nDuring extinction recall, the anterior pulvinar and MD exhibited similar functional patterns 267 \nat both block and trial levels, with increased activation to CS+ compared to CS - during 268 \nthe first two trials. In contrast, the medial and inferior pulvinar maintained the BOLD signal 269 \nto CS+ in the second trial while showing reduced responses to CS -. We observed no 270 \nactivation differences in the lateral pulvinar (Fig. 5b-c, Extended Tables 6-7). This finding 271 \nhighlights the involvement of most pulvinar divisions and MD in sustaining threat memory 272 \nunder safe contextual cues and suggests retrieval suppression of the extinguished threat.   273 \n 274 \nChanging the contextual cues to a threat background where the original associative 275 \nlearning occurred elicited increased activation in response to CS+ in the anterior, inferior, 276 \nand lateral pulvinar divisions along with the MD (Fig. 6b-c, Extended Tables 8-9). These 277 \nregions are, therefore, involved in the reactivation of threat memory, a hallmark process 278 \nin the threat renewal phase.  279 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 13 \n 280 \nConsistent LGN activation patterns across all threat learning phases align with 281 \nprinciples of feedforward processing. 282 \nAt the block level, we observed elevated activation in response to CS+ compared to CS- 283 \nacross conditioning, extinction, recall, and renewal. These differences were driven by the 284 \nfirst trial, during which an increased BOLD signal in response to CS+ was followed by a 285 \ndecline in subsequent trials ( Figs. 1d, 4d-6d, Extended Tables 2, 5, 7, and 9 ). These 286 \nfindings suggest that the LGN plays a consistent role across different phases of threat 287 \nlearning, potentially reflecting early attentional processing or anticipation of upcoming 288 \nvisual stimuli. 289 \n 290 \nThalamic connectivity underlies threat learning and memory  291 \nWe tested the relationships between each thalamic nucleus and core brain regions within 292 \nthe ‘fear circuit’(Herry et al., 2010; Maren and Quirk, 2004; Milad et al., 2014; Milad and 293 \nQuirk, 2012; Tovote et al., 2015)  by seeding the nuclei to target the amygdala, 294 \nhippocampus, ventromedial prefrontal cortex ( vmPFC), subgenual anterior cingulate 295 \ncortex (sgACC), and dorsal anterior cingulate cortex ( dACC). During conditioning, the 296 \nanterior pulvinar exhibited positive connectivity with the amygdala, vmPFC, and 297 \nhippocampus (all pFDR < 0.05 and all 95% bias -corrected and accelerated (BCa) 298 \nconfidence intervals (CI) of the mean differences from 10,000 bootstrap resamples 299 \nexcluded zero; Cohen’s ds = 0.14, 0.14, and 0.15, with corresponding 95% parametric 300 \nCIs also excluding zero ) (Fig. 1e). The connectivity with the amygdala likely supports 301 \nencoding the emotional valence of the newly learned  CS-US associations , w hile the 302 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 14 \nengagement of the hippocampus suggests prioritizing the contextual CS+ information. 303 \nThe connectivity with the vmPFC may underlie a process of top-down control release to 304 \nenhance fear expression or encode threat information for tracking and supporting 305 \ndecision-making in future encounters.  306 \nDuring extinction learning, we found increased LGN-sgACC connectivity (pFDR < 0.05 and 307 \n95% BCa CIs of the mean differences from 10,000 bootstrap resamples excluded zero ; 308 \nCohen’s d = 0.15, 95% parametric CI [0.035, 0.256] ), suggesting sgACC involvement in 309 \nemotional regulation ( Fig. 4e). Interestingly, i n recall and renewal, we found that the 310 \nlearners, i.e., the anterior pulvinar and MD  contribute to support ing either  safe or 311 \nthreatening memory , depending on contextual cues . Specifically,  the MD -dACC 312 \nconnectivity supported safe memory in recall (pFDR < 0.05 and 95% BCa CI of the mean 313 \ndifferences from 10,000 bootstrap resamples excluded zero ; Cohen’s d = 0.13, 95% 314 \nparametric CI [0.031, 0.225];  Fig. 5e), whereas anterior pulvinar -vmPFC supported 315 \nthreatening memory in renewal (p FDR < 0.01 and 95% BCa  CI of the mean differences 316 \nfrom 10,000 bootstrap resamples excluded zero ; Cohen’s d = 0. 20, 95% parametric CI 317 \n[0.084, 0.306];  Fig. 6e).  318 \nDiscussion 319 \nWe examined the neural representation of associative threat learning within the pulvinar 320 \ndivisions, LGN, and MD, providing new insights into thalamic involvement in this adaptive 321 \nbehavior in humans. We identified distinct roles among these thalamic nuclei with respect 322 \nto their activation profiles during threat learning and memory. The anterior pulvinar and 323 \nMD exhibited parallel activation patterns consistent with associative learning, reflecting 324 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 15 \ntheir roles in automatic survival responses and deliberate threat processing. We propose 325 \na novel hierarchical model for processing threat information in the pulvinar divisions. The 326 \nmedial pulvinar mediates basic sensory information from the inferior and lateral divisions 327 \nto the anterior pulvinar for higher-order integrative learning. Both pulvinar and MD were 328 \ninvolved in extinction and showed activation consistent with the salience processing of  329 \nthreat-related memories during extinction recall and threat renewal. The LGN primarily 330 \nrepresented feedforward processing, anticipating upcoming visual stimuli throughout all 331 \nphases of threat learning. We integrated these insights into schematic models underlying 332 \nthe emotional and behavioral expression of threat learning  and memory , providing a 333 \nneural framework for studying related human behaviors (Fig. 7a-b).  334 \nThe anterior pulvinar and MD  co-activation patterns demonstrate a parallel contribution 335 \nto threat learning. The similar activation in response to CS+ and CS - at the first trial, 336 \nfollowed by a heightened activation specific to CS+ in the next trial, is consistent with the 337 \nacquisition of the CS-US association. The responses’ similarity to both types of CS during 338 \nthe first trial suggests an equivalent initial emotional valence. The gradual increase in the 339 \nactivation, specifically to CS+ by the end of trial 1, suggests a shift in the emotional 340 \nvalence of CS+, indicating rapid associative learning. This parallel specialized role 341 \nhighlights the anterior pulvinar and MD as central  thalamic hubs for integrating CS -US 342 \ninformation, specifying the two-system model proposed by LeDoux and Pine(LeDoux and 343 \nPine, 2016) . Briefly, fear processing, a ccording to this model, involves two distinct 344 \npathways: a subcortical route for rapid threat detection and species -specific defense 345 \nreactions and a cortical route for deliberate threat evaluation and conscious fear 346 \nexperience. Learning within the anterior pulvinar appears to support rapid automatic 347 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 16 \nprocesses of the subcortical pathway. This aligns with previous reports that demonstrated 348 \nthe role of the pulvinar-amygdala pathway in encoding negative emotions  in 349 \nhumans(Bridge et al., 2015; Koller et al., 2019; Kragel et al., 2021; McFadyen et al., 2019; 350 \nRafal et al., 2015) . Indeed, the increased  connectivity that we observed with the 351 \namygdala, hippocampus, and vmPFC  during threat conditioning aligns with this role. 352 \nSpecifically, in this context, the amygdala  likely contributes to tagging a negative 353 \nemotional valence to CS+ and triggering fight, flight, or freeze reactions (Adolphs et al., 354 \n1994; Costa et al., 2022; Fanselow  and LeDoux, 1999; Phelps and LeDoux, 2005; Wen 355 \net al., 2024, 2022) . The connectivity with the hippocampus likely facilitates contextual 356 \nencoding of the environmental characteristics of the aversive event(Maren, 2001; Maren 357 \net al., 2013). In turn, the connectivity with vmPFC appears to support salience encoding 358 \nfor tracking future encounters with the learned threat(Battaglia et al., 2020) and facilitating 359 \ndecision-making and emotion regulation(Milad and Quirk, 2012, 2002; Nejati et al., 2021).  360 \nOn the other hand, MD activation aligns more closely with the cortical pathway, facilitating 361 \nconscious and deliberate threat encoding. This is supported by our findings, which 362 \nshowed increased MD activation in response to CS+ , compared to the anterior pulvinar  363 \nduring threat learning , i ndicating a broader or deeper processing of the CS -US 364 \nassociation. Additionally, this interpretation is supported by substantial evidence pointing 365 \nto well -established anatomical pathways connecting the MD with the PFC and 366 \ndemonstrating its role in  decision-making and learning (Behrens et al., 2003; Hwang et 367 \nal., 2020; Li et al., 2022; Wolff and Halassa, 2024) . This is a long with reports that 368 \nunderscored the  contribution of the  MD-cortical loops to supporting a conscious 369 \nexperience in humans(Griffiths et al., 2022; Whyte et al., 2024).  Together, these findings 370 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 17 \nadvance our understanding of the two-system model of threat processing, highlighting the 371 \nanterior pulvinar and MD role in the acquisition of the CS-US association and suggesting 372 \nthat these regions underlie a parallel conscious vs. unconscious learning proposed by 373 \nLeDoux and colleagues(LeDoux and Pine, 2016), Fig. 7a.   374 \nOur findings demonstrate that different pulvinar divisions hierarchically contribute to the 375 \nintegration of CS-US association in the anterior pulvinar through bottom -up processing. 376 \nAlthough the anatomical projections within pulvinar divisions are not well characterized, 377 \nour functional data-driven approach revealed a critical role for the medial pulvinar as a 378 \nhub region facilitating communication among pulvinar divisions. Specifically, our trial-wise 379 \nanalysis showed that the medial, inferior, and lateral pulvinar process threat -related 380 \ninformation earlier than the anterior pulvinar. The anterior pulvinar exhibited activation in 381 \nresponse to CS+ starting at trial 2, whereas activation in the other pulvinar divisions 382 \noccurred as early as trial 1. Network modeling further identified the medial pulvinar as a 383 \ncentral hub, facilitating communication with other  divisions. Using a robust mediation 384 \nmodel, we demonstrated that the medial pulvinar appears to mediate the relay of sensory 385 \nCS+ information from the inferior and lateral pulvinar to the anterior division for higher-386 \norder integra tive learning . This hierarchical model aligns with previous evidence 387 \nsuggesting that the inferior and lateral pulvinar are associated with  processing basic 388 \nsensory information(Berman and Wurtz, 2010, 2008; Bridge et al., 2015; Cortes et al., 389 \n2024). W hile the medial pulvinar , which receives projections from deep layers of the 390 \nsuperior colliculus, supports more advanced functions such as attention and working 391 \nmemory(Bridge et al., 2015; Homman-Ludiye and Bourne, 2019). These findings provide 392 \nnovel insights into the functional specialization of pulvinar divisions during threat learning, 393 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 18 \nsuggesting feedforward functions in the inferior and lateral pulvinar and high -order 394 \nintegration in the anterior pulvinar.   395 \nThe pulvinar divisions and MD are actively engaged in extinction learning, as indicated 396 \nby increased activation in response to CS+ during the first two trials, followed by a 397 \ndiminished response. This decline in activation likely reflects a shift in the emotional 398 \nvalence of the CS+ to match that of the CS-, suggesting a rapid extinction process in the 399 \nhuman thalamus. Despite the involvement in extinction, these nuclei remained sensitive 400 \nto the extinguished threat during safe contextual cues in extinction recall and exhibited 401 \nincreased activation  to the extinguished CS+ when the stimuli were presented with in 402 \nthreat contextual cues during renewal. This pattern suggests engagement in retrieval 403 \nsuppression during recall , along with  threat salience processing during both recall and 404 \nrenewal. The MD and pulvinar continue to monitor and evaluate the extinguished threat 405 \nacross contexts rather than relying on static safety memories . These nuclei appear to 406 \nengage in dynamic evaluation, learning, and decision -making during future encounters 407 \nwith the extinguished CS+. 408 \nThe classical Pavlovian model  proposes that extinction learning forms a new safety 409 \nmemory, competing with the original threat memory acquired during conditioning(Bouton, 410 \n2002; Milad and Quirk, 2002) . Contextual changes often favor either the safe or threat 411 \nmemory. Maren and colleagues (Maren et al., 2013; Maren and Quirk, 2004)  described 412 \nthe hippocampal-prefrontal-amygdala model for contextual memory. The hippocampus 413 \nprojects to the basolateral amygdala, vmPFC, and dACC. Although the vmPFC supports 414 \nsafe memory recall by projecting to the intercalated cells (which inhibit the central nucleus 415 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 19 \nof the amygdala ), the dACC enhances threat memory renewal by projecting to the 416 \nbasolateral amygdala (which activates the central nucleus of this structure). Our findings 417 \nsuggest that the thalamic connectivity may influence the balance between recalling safe 418 \nvs. threat memory by  modulating the competitive interaction between the dACC and 419 \nvmPFC. The increased MD-dACC connectivity during recall may reflect a thalamic-driven 420 \nreconfiguration of prefrontal circuits, enabling the vmPFC to become more functionally 421 \ndominant and promote safe memory expression. On the other hand, the anterior pulvinar-422 \nvmPFC connectivity during renewal may destabilize the influence of the vmPFC, allowing 423 \nthe dACC to regain dominance and facilitate threat memor y retrieval . The thalamic 424 \nconnectivity, thus,  appears to orchestrate a context-dependent functional balance  425 \nbetween the dACC and vmPFC. This balance may serve as a neural \"toggle switch\" 426 \nbetween safe vs. threat memory expression. We integrated this  suggested flexible 427 \nresponding mechanism to changing environmental contexts  in Maren’s circuit model of 428 \nemotional memory, Fig. 7b. 429 \nThe LGN contribution is consistent across all phases of threat learning, as evidenced by 430 \nour trial -wise results, which pointed to increased activation in response to CS+ that 431 \ndiminished immediately after the first trial. This distinct pattern indicates readiness for 432 \nimminent visual stimuli, regardless of the actual emotional valence, as conditioning, 433 \nextinction, recall, and renewal elicited similar BOLD patterns. This aligns with previous 434 \nstudies that highlighted the LGN’s engagement in selective attention and anticipation of 435 \nvisual stimuli (Mahoney and Schmidt, 2024; O’Connor et al., 2002; Saalmann and 436 \nKastner, 2009). The LGN, as a first-order nucleus(Cortes et al., 2024),  primarily supports 437 \na feedforward function, while the pulvinar, as part of the visual thalamus, is more closely 438 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 20 \nassociated with a broader functional processing(Cortes et al., 2024) including emotional 439 \nvalences of stimuli. This distinction underscores their joint but specialized contributions 440 \nto adaptive threat responses. 441 \nAlthough distinct thalamic roles in threat learning have been proposed, fMRI data do not 442 \nfully capture the complexity of this structure. Pulvinar divisions, MD, and LGN each 443 \ncontain different neuron subtypes and f iner anatomical subdivisions, which may serve 444 \ndiverse functions. Future advancements, such as higher-resolution human brain atlases, 445 \ncould improve our ability to study these nuclei with better anatomical precision. 446 \nAdditionally, since different sensory modalities preferentially engage distinct thalamic 447 \nnuclei, the specific thalamic roles we described may not be consistent across different 448 \nexperimental designs, particularly in studies using auditory threat learning. The pulvinar 449 \ndivisions’ relationships during conditioning are purely functional and might be supported 450 \nby direct or indirect anatomical projections. Finally, given the indirect nature of fMRI data 451 \nand the absence of direct brain signal manipulations, our findings should not be 452 \ninterpreted as evidence of causality. Further research is needed to examine causal 453 \nmechanisms underlying dynamic neural representations of threat learning within the 454 \nthalamus.  455 \nThis study’s  insights raise critical future questions at the intersection between 456 \nneuroscience, psychology, and mental health , particularly regarding the neural 457 \nmechanisms underlying associative threat learning . First, the crucial role of the anterior 458 \npulvinar and MD during the acquisition of the CS-US association sets the stage for 459 \ndeveloping more precise brain interventions, focusing on  refining maladaptive fear 460 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 21 \nreactions. Future studies could explore the optimal parameters for applying non-invasive 461 \ntechniques to inhibit these thalamic regions during or immediately after fear conditioning. 462 \nThis approach is particularly promising for clinical populations and individuals exposed to 463 \nhigh-risk environments, such as emergency medical staff, firefighters, and paramedics.  464 \nSecond, the distinct contributions of the MD -dACC and anterior pulvinar-vmPFC circuits 465 \nto safe vs. threatening memory suggest potential intervention targets for prioritizing one 466 \nmemory over the other . Activating the MD -dACC may facilitate recall ing safe memories 467 \nwhile engaging the anterior pulvinar-vmPFC circuits might reinforce fear memory 468 \npathways. Experimental designs in controlled laboratory settings could target these 469 \ncircuits to identify specific parameters for enhancing safe memor ies. This avenue holds 470 \ntherapeutic potential , particularly in conditions such as aviophobia ; activating a safe 471 \nmemory circuit before a flight might suppress fear relapse during the actual flight.  472 \nThird, the anterior pulvinar-MD relationships raise questions related to the wide range of 473 \nhuman behaviors, focusing on understanding neural gateways between conscious and 474 \nunconscious learning. Studying the dynamics of information flow across this promising 475 \npathway could deepen our understanding of how sensory and cognitive information 476 \ntransits between conscious and unconscious states  and bring s about behaviors . This 477 \nresearch line  may pave the way for innovative learning methods and  reveal 478 \nneurocognitive mechanisms underlying the acquisition of new information in humans.  479 \nTogether, our findings and emerging research directions underscore the thalamic nuclei’s 480 \nvital role as a hub for fear acquisition and memory processing, offering promising avenues 481 \nfor theoretical advancements and clinical applications. As the primary neural gateway to 482 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 22 \nthe human brain, the thalamus is the first station for all sensory information except 483 \nolfactory input s. Modulating its functions  is potentially an impactful strategy to induce 484 \nwidespread functional changes across the brain and influence different mental and  485 \nbehavioral expressions.  486 \nMaterials and Methods  487 \nParticipants  488 \nWe analyzed the fMRI data of 293-412 human subjects of both sexes, aged 18-70 years 489 \nold (M = 32.17 ± 13.1). All participants were proficient in English, right -handed, and had 490 \nnormal or corrected-to-normal vision. The exclusion criteria included a history of seizures 491 \nor significant head trauma, current substance abuse or dependence, metal implants, 492 \npregnancy, breastfeeding, or positive urine toxicology screen for drugs of abuse. We 493 \nfollowed the latest version of Helsinki's declaration, and all procedures were approved by 494 \nthe Partners HealthCare Institute Review Board of the Massachusetts General Hospital, 495 \nHarvard Medical School. All subjects provided written informed consent before taking part 496 \nin the study. Results from this dataset have been published elsewhere with a different 497 \nfocus(Marin et al., 2020; Wen et al., 2024, 2022). The current results are novel and have 498 \nnot been previously published.  499 \n 500 \nExperimental procedure 501 \n  502 \nParticipants underwent two -day sessions of a validated threat learning paradigm in the 503 \nMRI scanner (Figs. 1a, 4a-6a); the experimental contexts consisted of visual scenes on 504 \na computer display. On day 1, participants underwent a Pavlovian conditioning phase in 505 \nwhich a neutral stimulus in context A (e.g., red light in an office) was paired with a 500ms 506 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 23 \nelectric shock (US) with a partial reinforcement rate of 62.5%. Another neutral stimulus in 507 \ncontext A (e.g., blue light in the same office) was also presented but never paired with the 508 \nshock (CS-). Participants were guided to select a highly annoying but nonpainful shock 509 \nlevel during a pre -experiment calibration stage, and we used that personalized level 510 \nduring conditioning. On the same day, we also conducted extinction learning in context B 511 \n(e.g., red and blue lights in a casebook background).  The stimuli were repeatedly 512 \npresented with the removal of the expected reinforcement (i.e., no shock).  513 \n 514 \nOn day 2, the participants underwent two phases of a memory test. The first is extinction 515 \nrecall, including presenting the extinguished CS+ and CS - within safe contextual cues 516 \n(context B used during extinction learning). The second is  threat renewal, in which the 517 \nstimuli were presented within threat contextual cues (context A used in conditioning).  518 \nBoth extinction recall and threat renewal included no US reinforcement. Across phases 519 \nof threat learning , the duration of each trial was 6s, and the inter -trial intervals with a 520 \nfixation screen ranged between 12 to 18  s (15  s on average). Finally, to control for 521 \npotential confounds related to the experimental design, we pseudorandomized and 522 \ncounterbalanced the order of CS+ and CS− across phases and between subjects. 523 \n 524 \nMRI data acquisition and preprocessing  525 \nTwo MRI settings were used to acquire the neuroimaging data. The first is a Trio 3T whole-526 \nbody MRI scanner (Siemens Medical Systems, Iselin, New Jersey) using an 8 -channel 527 \nhead coil. The functional data in this setting were acquired using a T2* weighted echo -528 \nplanar pulse sequence with these parameters: TR = 3.0s, TE = 30 ms, slice number = 45, 529 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 24 \nvoxel size = 3 × 3 × 3 mm3. The second setting was also in the same scanner using a 32-530 \nchannel head coil. The functional images were obtained using a T2* weighted echo-planar 531 \npulse sequence using TR = 2.56s TE  = 30 ms, slice number  = 48, voxel 532 \nsize = 3 × 3 × 3 mm3. The anatomical brain images were collected using a T1 -weighted 533 \nMP-RAGE pulse sequence , parcellated into 1  × 1 × 1 mm3 voxels. Elastic bands were 534 \naffixed to the head coil device to reduce head motions. 535 \n 536 \nUsing the default pipeline in fMRIPrep, version 20.0.2 (Esteban et al., 2020, 2019) , we 537 \npreprocessed the data and applied correction of slice timing, realignment of the functional 538 \nimages, and coregistration. In addition, the data was normalized into the Montreal 539 \nNeurological Institute (MNI) space and smoothed with a 6 -mm full-width half-maximum 540 \nGaussian kernel. 541 \n 542 \nActivation analyses  543 \nWe applied the least-squares-based generalized linear model (GLM) for each participant 544 \nto estimate the BOLD response to CS+ and CS - using SPM 12. We estimated the beta 545 \nvalues for each voxel during  each learning phase in  the paradigm. Overall, the model 546 \nincluded 32 regressors for the CS+ and CS-, a regressor for the context, and a regressor 547 \nfor shock in conditioning but not in other phases . The GLM also included six head 548 \nmovement parameters (x, y, z directions, and rotations). This first-level analysis resulted 549 \nin contrast maps that we used to estimate the variability of these maps across all subjects 550 \nat the group-level analysis. We then used the contrast maps from the group-level analysis 551 \nto extract the averaged values across the voxels within predefined masks of the pulvinar 552 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 25 \ndivisions, MD, and LGN. These outputs were used to compare the BOLD response during 553 \nthe first four CS+ and four CS- trials. We averaged the activation for each CS type across 554 \ntrials to obtain the block-level activation.  555 \n 556 \nThe statistical analyses included t-tests, RM -ANOVA, network analyses, mediation 557 \nmodels, and hierarchical regression  models. All were conducted using JASP versions 558 \n0.18.3 and 0.19.3(JASP Team, 2024). False discovery rate (FDR) correction was applied 559 \nacross analyses. RM-ANOVA was used to analyze trial-by-trial BOLD responses as the 560 \nsame participants were measured across the trials. Assumptions of sphericity were 561 \nchecked, and Greenhouse-Geisser corrections were applied where necessary.  562 \n 563 \nConnectivity analyses  564 \nWe computed the connectivity values  using the CONN functional connectivity toolbox , 565 \nversion 22. a, for the MathWorks MATLAB program (Nieto-Castanon and Whitfield -566 \nGabrieli, 2022; Whitfield -Gabrieli and Nieto -Castanon, 2012) . We first segmented the 567 \nbrain images into different tissues of gray matter, white matter, and CSF. Then, we applied 568 \nthe standard denoising pipeline in the CONN toolbox (Nieto-Castanon, 2020)  to the 569 \nfunctional images to control the effect of potential confounding parameters, using  a 570 \ncomponent-based noise correction method . Finally, bandpass frequency filtering of the 571 \nBOLD time series between 0.008 Hz and 0.09 Hz was also applied.  572 \nWe evaluated differences in connectivity between CS+ and CS− conditions using 573 \ngeneralized psychophysiological interaction (gPPI) analyses.  We defined the first eight 574 \ntrials of each condition as a block. The pulvinar divisions, MD, and LGN were defined as 575 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 26 \nindividual seed regions, and connectivity was assessed with target regions, including the 576 \ndACC, sgACC, vmPFC, amygdala, and hippocampus.  The model included seed BOLD 577 \nsignals as physiological factors, boxcar signals characterizing the task conditions 578 \nconvolved with an SPM canonical hemodynamic response function as psychological 579 \nfactors, and the product of the two as psychophysiological interaction terms. Functional 580 \nconnectivity changes were quantified using Fisher-transformed correlation coefficients of 581 \nthe psychophysiological interaction terms. At the group level, we used a GLM to assess 582 \ntask-related connectivity changes across participants. Differences in connectivity 583 \nbetween conditions were evaluated using paired t -tests, applying false discovery rate 584 \n(FDR) correction at p < 0.05 . 10,000 bootstrap resamples were used to obtain BCa CI , 585 \nproviding more robust estimates of the connectivity mean differences.  586 \nMasks 587 \nWe defined the pulvinar divisions, MD, and LGN nuclei using predefined masks based on 588 \nthe neuroanatomical guidelines of the Automated Anatomical Labelling Atlas(Rolls et al., 589 \n2020). We also applied the same atlas guidelines to determine the masks for the  two 590 \nother anatomical regions; the amygdala and hippocampus. The masks for the functional 591 \nregions were created using Neurosynth (Yarkoni et al., 2011)  and the keyword 592 \n‘conditioning.’ For each region, we created 8mm spheres around the following identified 593 \npeak coordinates: vmPFC (MNIxyz  = −2, 46, −10), sgACC (MNIxyz  = 0, 26, −12), and 594 \ndACC (MNIxyz = 0, 14, 28). 595 \n 596 \n 597 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 27 \nAcknowledgments  598 \nThis work was supported by the National Institute of Mental Health grants 599 \nR01MH123736, R01MH125198, R33MH111907, R01MH097880, and R01MH097964 to 600 \nM.R.M. 601 \n 602 \n 603 \n 604 \n 605 \n 606 \n 607 \n 608 \n 609 \n 610 \n 611 \n 612 \n 613 \n 614 \n 615 \n 616 \n 617 \n 618 \n 619 \n 620 \n 621 \n 622 \n 623 \n 624 \n 625 \n 626 \n 627 \n 628 \n 629 \n 630 \n 631 \n 632 \n 633 \n 634 \n 635 \n 636 \n 637 \n 638 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 28 \nReferences  639 \nAdolphs R, Tranel D, Damasio H, Damasio A. 1994. Impaired recognition of emotion in 640 \nfacial expressions following bilateral damage to the human amygdala. Nature 1994 641 \n372:6507 372:669–672. doi:10.1038/372669a0 642 \nArcaro MJ, Pinsk MA, Kastner S. 2015. The Anatomical and Functional Organization of 643 \nthe Human Visual Pulvinar. Journal of Neuroscience 35:9848–9871. 644 \ndoi:10.1523/JNEUROSCI.1575-14.2015 645 \nBadarnee M, Wen Z, Hammoud MZ, Glimcher P, Cain CK, Milad MR. 2025. Intersect 646 \nbetween brain mechanisms of conditioned threat, active avoidance, and reward. 647 \nCommunications Psychology 3:32. doi:10.1038/S44271-025-00197-7 648 \nBattaglia S, Garofalo S, di Pellegrino G, Starita F. 2020. Revaluing the Role of vmPFC 649 \nin the Acquisition of Pavlovian Threat Conditioning in Humans. Journal of 650 \nNeuroscience 40:8491–8500. doi:10.1523/JNEUROSCI.0304-20.2020 651 \nBehrens TEJ, Johansen-Berg H, Woolrich MW, Smith SM, Wheeler-Kingshott CAM, 652 \nBoulby PA, Barker GJ, Sillery EL, Sheehan K, Ciccarelli O, Thompson AJ, Brady 653 \nJM, Matthews PM. 2003. Non-invasive mapping of connections between human 654 \nthalamus and cortex using diffusion imaging. Nat Neurosci 6:750–757. 655 \ndoi:10.1038/NN1075;KWRD=BIOMEDICINE 656 \nBerman RA, Wurtz RH. 2010. Functional Identification of a Pulvinar Path from Superior 657 \nColliculus to Cortical Area MT. Journal of Neuroscience 30:6342–6354. 658 \ndoi:10.1523/JNEUROSCI.6176-09.2010 659 \nBerman RA, Wurtz RH. 2008. EXPLORING THE PULVINAR PATH TO VISUAL 660 \nCORTEX. Prog Brain Res 171:467. doi:10.1016/S0079-6123(08)00668-7 661 \nBertini C, Pietrelli M, Braghittoni D, Làdavas E. 2018. Pulvinar lesions disrupt fear-662 \nrelated implicit visual processing in hemianopic patients. Front Psychol 9:412050. 663 \ndoi:10.3389/FPSYG.2018.02329/BIBTEX 664 \nBouton ME. 2002. Context, ambiguity, and unlearning: Sources of relapse after 665 \nbehavioral extinction. Biol Psychiatry 52:976–986. doi:10.1016/S0006-666 \n3223(02)01546-9 667 \nBridge H, Leopold DA, Bourne JA. 2015. Adaptive pulvinar circuitry supports visual 668 \ncognition. Trends Cogn Sci 20:146. doi:10.1016/J.TICS.2015.10.003 669 \nCarr JA. 2015. I’ll take the low road: The evolutionary underpinnings of visually triggered 670 \nfear. Front Neurosci 9:165437. doi:10.3389/FNINS.2015.00414/BIBTEX 671 \nCasanova C, Chalupa LM. 2023. The dorsal lateral geniculate nucleus and the pulvinar 672 \nas essential partners for visual cortical functions. Front Neurosci 17:1258393. 673 \ndoi:10.3389/FNINS.2023.1258393/BIBTEX 674 \nCortes N, Ladret HJ, Abbas-Farishta R, Casanova C. 2024. The pulvinar as a hub of 675 \nvisual processing and cortical integration. Trends Neurosci 47:120–134. 676 \ndoi:10.1016/J.TINS.2023.11.008 677 \nCosta M, Lozano-Soldevilla D, Gil-Nagel A, Toledano R, Oehrn CR, Kunz L, Yebra M, 678 \nMendez-Bertolo C, Stieglitz L, Sarnthein J, Axmacher N, Moratti S, Strange BA. 679 \n2022. Aversive memory formation in humans involves an amygdala-hippocampus 680 \nphase code. Nature Communications 2022 13:1 13:1–16. doi:10.1038/s41467-022-681 \n33828-2 682 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 29 \nEsteban O, Ciric R, Finc K, Blair RW, Markiewicz CJ, Moodie CA, Kent JD, Goncalves 683 \nM, DuPre E, Gomez DEP, Ye Z, Salo T, Valabregue R, Amlien IK, Liem F, Jacoby N, 684 \nStojić H, Cieslak M, Urchs S, Halchenko YO, Ghosh SS, De La Vega A, Yarkoni T, 685 \nWright J, Thompson WH, Poldrack RA, Gorgolewski KJ. 2020. Analysis of task-686 \nbased functional MRI data preprocessed with fMRIPrep. Nat Protoc 15:2186–2202. 687 \ndoi:10.1038/S41596-020-0327-3 688 \nEsteban O, Markiewicz CJ, Blair RW, Moodie CA, Isik AI, Erramuzpe A, Kent JD, 689 \nGoncalves M, DuPre E, Snyder M, Oya H, Ghosh SS, Wright J, Durnez J, Poldrack 690 \nRA, Gorgolewski KJ. 2019. fMRIPrep: a robust preprocessing pipeline for functional 691 \nMRI. Nat Methods 16:111–116. doi:10.1038/S41592-018-0235-4 692 \nFanselow MS, LeDoux JE. 1999. Why we think plasticity underlying pavlovian fear 693 \nconditioning occurs in the basolateral amygdala. Neuron 23:229–232. 694 \ndoi:10.1016/S0896-6273(00)80775-8 695 \nFoygel R, Drton M. 2010. Extended Bayesian Information Criteria for Gaussian 696 \nGraphical Models. Adv Neural Inf Process Syst 23. 697 \nFullana MA, Harrison BJ, Soriano-Mas C, Vervliet B, Cardoner N, Àvila-Parcet A, Radua 698 \nJ. 2015. Neural signatures of human fear conditioning: an updated and extended 699 \nmeta-analysis of fMRI studies. Molecular Psychiatry 2016 21:4 21:500–508. 700 \ndoi:10.1038/mp.2015.88 701 \nGriffiths BJ, Zaehle T, Repplinger S, Schmitt FC, Voges J, Hanslmayr S, Staudigl T. 702 \n2022. Rhythmic interactions between the mediodorsal thalamus and prefrontal 703 \ncortex precede human visual perception. Nature Communications 2022 13:1 13:1–704 \n11. doi:10.1038/s41467-022-31407-z 705 \nHalassa MM, Sherman SM. 2019. Thalamo-cortical circuit motifs: a general framework. 706 \nNeuron 103:762. doi:10.1016/J.NEURON.2019.06.005 707 \nHalverson HE, Freeman JH. 2010. Ventral lateral geniculate input to the medial pons is 708 \nnecessary for visual eyeblink conditioning in rats. Learn Mem 17:80–85. 709 \ndoi:10.1101/LM.1572710 710 \nHerry C, Ferraguti F, Singewald N, Letzkus JJ, Ehrlich I, Lüthi A. 2010. Neuronal circuits 711 \nof fear extinction. European Journal of Neuroscience 31:599–612. 712 \ndoi:10.1111/J.1460-9568.2010.07101.X 713 \nHomman-Ludiye J, Bourne JA. 2019. The medial pulvinar: function, origin and 714 \nassociation with neurodevelopmental disorders. J Anat 235:507–520. 715 \ndoi:10.1111/JOA.12932 716 \nHummos A, Wang BA, Drammis S, Halassa MM, Pleger B. 2022. Thalamic regulation of 717 \nfrontal interactions in human cognitive flexibility. PLoS Comput Biol 18:e1010500. 718 \ndoi:10.1371/JOURNAL.PCBI.1010500 719 \nHwang K, Bertolero MA, Liu WB, D’Esposito M. 2017. The Human Thalamus Is an 720 \nIntegrative Hub for Functional Brain Networks. Journal of Neuroscience 37:5594–721 \n5607. doi:10.1523/JNEUROSCI.0067-17.2017 722 \nHwang K, Bruss J, Tranel D, Boes AD. 2020. Network Localization of Executive 723 \nFunction Deficits in Patients with Focal Thalamic Lesions. J Cogn Neurosci 724 \n32:2303. doi:10.1162/JOCN_A_01628 725 \nJASP Team. 2024. JASP (Versions 0.18.3 and 0.19.3). 726 \nKoller K, Rafal RD, Platt A, Mitchell ND. 2019. Orienting toward threat: Contributions of 727 \na subcortical pathway transmitting retinal afferents to the amygdala via the superior 728 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 30 \ncolliculus and pulvinar. Neuropsychologia 128:78–86. 729 \ndoi:10.1016/J.NEUROPSYCHOLOGIA.2018.01.027 730 \nKolling N, Wittmann M, Rushworth MFS. 2014. Multiple neural mechanisms of decision 731 \nmaking and their competition under changing risk pressure. Neuron 81:1190–1202. 732 \ndoi:10.1016/J.NEURON.2014.01.033/ATTACHMENT/A9984A25-8961-453B-8E1A-733 \n778EAD3A0ADE/MMC1.PDF 734 \nKonrad C, Neuhoff L, Adolph D, Goerigk S, Herbert JS, Jagusch-Poirier J, Weigelt S, 735 \nSeehagen S, Schneider S. 2024. Associative learning via eyeblink conditioning 736 \ndiffers by age from infancy to adulthood. Communications Psychology 2024 2:1 737 \n2:1–13. doi:10.1038/s44271-024-00176-4 738 \nKorn CW, Bach DR. 2019. Minimizing threat via heuristic and optimal policies recruits 739 \nhippocampus and medial prefrontal cortex. Nature Human Behaviour 2019 3:7 740 \n3:733–745. doi:10.1038/s41562-019-0603-9 741 \nKorn CW, Bach DR. 2018. Heuristic and optimal policy computations in the human brain 742 \nduring sequential decision-making. Nature Communications 2018 9:1 9:1–15. 743 \ndoi:10.1038/s41467-017-02750-3 744 \nKragel PA, Čeko M, Theriault J, Chen D, Satpute AB, Wald LW, Lindquist MA, Feldman 745 \nBarrett L, Wager TD. 2021. A human colliculus-pulvinar-amygdala pathway 746 \nencodes negative emotion. Neuron 109:2404-2412.e5. 747 \ndoi:10.1016/J.NEURON.2021.06.001/ATTACHMENT/2E7063C2-7363-4DE3-9A2B-748 \nF75C55D9DD96/MMC3.PDF 749 \nKrasne FB, Zinn R, Vissel B, Fanselow MS, Franklin Krasne CB, Howland-Rose V, 750 \nBattersby D, King D. 2021. Extinction and discrimination in a Bayesian model of 751 \ncontext fear conditioning (BaconX). Hippocampus 31:790–814. 752 \ndoi:10.1002/HIPO.23298 753 \nLe Q Van, Isbell LA, Matsumoto J, Nguyen M, Hori E, Maior RS, Tomaz C, Tran AH, 754 \nOno T, Nishijo H. 2013. Pulvinar neurons reveal neurobiological evidence of past 755 \nselection for rapid detection of snakes. Proc Natl Acad Sci U S A 110:19000–756 \n19005. doi:10.1073/PNAS.1312648110/SUPPL_FILE/PNAS.201312648SI.PDF 757 \nLeDoux JE, Pine DS. 2016. Using Neuroscience to Help Understand Fear and Anxiety: 758 \nA Two-System Framework. https://doi.org/101176/appi.ajp201616030353 759 \n173:1083–1093. doi:10.1176/APPI.AJP.2016.16030353 760 \nLee S, Shin HS. 2016. The role of mediodorsal thalamic nucleus in fear extinction. J 761 \nAnal Sci Technol 7:1–5. doi:10.1186/S40543-016-0093-6/FIGURES/1 762 \nLee Sukchan, Ahmed T, Lee Soojung, Kim H, Choi S, Kim DS, Kim SJ, Cho J, Shin HS. 763 \n2011. Bidirectional modulation of fear extinction by mediodorsal thalamic firing in 764 \nmice. Nature Neuroscience 2011 15:2 15:308–314. doi:10.1038/nn.2999 765 \nLi K, Fan L, Cui Y, Wei X, He Y, Yang J, Lu Y, Li W, Shi W, Cao L, Cheng L, Li A, You B, 766 \nJiang T. 2022. The human mediodorsal thalamus: Organization, connectivity, and 767 \nfunction. Neuroimage 249:118876. doi:10.1016/J.NEUROIMAGE.2022.118876 768 \nLi XB, Inoue T, Nakagawa S, Koyama T. 2004. Effect of mediodorsal thalamic nucleus 769 \nlesion on contextual fear conditioning in rats. Brain Res 1008:261–272. 770 \ndoi:10.1016/J.BRAINRES.2004.02.038 771 \nLithari C, Moratti S, Weisz N. 2015. Thalamocortical interactions underlying visual fear 772 \nconditioning in humans. Hum Brain Mapp 36:4592. doi:10.1002/HBM.22940 773 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 31 \nMahoney HL, Schmidt TM. 2024. The cognitive impact of light: illuminating ipRGC circuit 774 \nmechanisms. Nature Reviews Neuroscience 2024 25:3 25:159–175. 775 \ndoi:10.1038/s41583-023-00788-5 776 \nMaren S. 2001. Nuerobiology of Pavlovian fear conditioning. Annu Rev Neurosci 777 \n24:897–931. doi:10.1146/ANNUREV.NEURO.24.1.897/CITE/REFWORKS 778 \nMaren S, Phan KL, Liberzon I. 2013. The contextual brain: implications for fear 779 \nconditioning, extinction and psychopathology. Nature Reviews Neuroscience 2013 780 \n14:6 14:417–428. doi:10.1038/nrn3492 781 \nMaren S, Quirk GJ. 2004. Neuronal signalling of fear memory. Nature Reviews 782 \nNeuroscience 2004 5:11 5:844–852. doi:10.1038/nrn1535 783 \nMarin MF, Hammoud MZ, Klumpp H, Simon NM, Milad MR. 2020. Multimodal 784 \nCategorical and Dimensional Approaches to Understanding Threat Conditioning 785 \nand Its Extinction in Individuals With Anxiety Disorders. JAMA Psychiatry 77:618–786 \n627. doi:10.1001/JAMAPSYCHIATRY.2019.4833 787 \nMcFadyen J, Mattingley JB, Garrido MI. 2019. An afferent white matter pathway from 788 \nthe pulvinar to the amygdala facilitates fear recognition. Elife 8. 789 \ndoi:10.7554/ELIFE.40766 790 \nMilad MR, Furtak SC, Greenberg JL, Keshaviah A, Im JJ, Falkenstein MJ, Jenike M, 791 \nRauch SL, Wilhelm S. 2013. Deficits in Conditioned Fear Extinction in Obsessive-792 \nCompulsive Disorder and Neurobiological Changes in the Fear Circuit. JAMA 793 \nPsychiatry 70:608–618. doi:10.1001/JAMAPSYCHIATRY.2013.914 794 \nMilad MR, Pitman RK, Ellis CB, Gold AL, Shin LM, Lasko NB, Zeidan MA, Handwerger 795 \nK, Orr SP, Rauch SL. 2009. Neurobiological Basis of Failure to Recall Extinction 796 \nMemory in Posttraumatic Stress Disorder. Biol Psychiatry 66:1075–1082. 797 \ndoi:10.1016/J.BIOPSYCH.2009.06.026 798 \nMilad MR, Quirk GJ. 2012. Fear Extinction as a Model for Translational Neuroscience: 799 \nTen Years of Progress. Annu Rev Psychol 63:129. 800 \ndoi:10.1146/ANNUREV.PSYCH.121208.131631 801 \nMilad MR, Quirk GJ. 2002. Neurons in medial prefrontal cortex signal memory for fear 802 \nextinction. Nature 2002 420:6911 420:70–74. doi:10.1038/nature01138 803 \nMilad MR, Rosenbaum BL, Simon NM. 2014. Neuroscience of fear extinction: 804 \nImplications for assessment and treatment of fear-based and anxiety related 805 \ndisorders. Behaviour Research and Therapy 62:17–23. 806 \ndoi:10.1016/J.BRAT.2014.08.006 807 \nMilad MR, Wright CI, Orr SP, Pitman RK, Quirk GJ, Rauch SL. 2007. Recall of fear 808 \nextinction in humans activates the ventromedial prefrontal cortex and hippocampus 809 \nin concert. Biol Psychiatry 62:446–454. doi:10.1016/J.BIOPSYCH.2006.10.011 810 \nMukherjee A, Lam NH, Wimmer RD, Halassa MM. 2021. Thalamic circuits for 811 \nindependent control of prefrontal signal and noise. Nature 2021 600:7887 600:100–812 \n104. doi:10.1038/s41586-021-04056-3 813 \nNejati V, Majdi R, Salehinejad MA, Nitsche MA. 2021. The role of dorsolateral and 814 \nventromedial prefrontal cortex in the processing of emotional dimensions. Scientific 815 \nReports 2021 11:1 11:1–12. doi:10.1038/s41598-021-81454-7 816 \nNieto-Castanon A. 2020. FMRI denoising pipeline. Handbook of functional connectivity 817 \nMagnetic Resonance Imaging methods in CONN 17–25. 818 \ndoi:10.56441/HILBERTPRESS.2207.6600 819 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 32 \nNieto-Castanon A, Whitfield-Gabrieli S. 2022. CONN functional connectivity toolbox: 820 \nRRID SCR_009550, release 22. CONN functional connectivity toolbox: RRID 821 \nSCR_009550, release 22. doi:10.56441/HILBERTPRESS.2246.5840 822 \nO’Connor DH, Fukui MM, Pinsk MA, Kastner S. 2002. Attention modulates responses in 823 \nthe human lateral geniculate nucleus. Nature Neuroscience 2002 5:11 5:1203–824 \n1209. doi:10.1038/nn957 825 \nPavlov I. 1904. Nobel Lecture. NobelPrize.org. Available at: 826 \nhttps://www.nobelprize.org/prizes/medicine/1904/pavlov/lecture/ 827 \nPaydar A, Lee B, Gangadharan G, Lee S, Hwang EM, Shin HS. 2014. Extrasynaptic 828 \nGABAA receptors in mediodorsal thalamic nucleus modulate fear extinction 829 \nlearning. Mol Brain 7:39. doi:10.1186/1756-6606-7-39 830 \nPenzo MA, Robert V, Tucciarone J, De Bundel D, Wang M, Van Aelst L, Darvas M, 831 \nParada LF, Palmiter RD, He M, Huang ZJ, Li B. 2015. The paraventricular thalamus 832 \ncontrols a central amygdala fear circuit. Nature 519:455–459. 833 \ndoi:10.1038/NATURE13978 834 \nPessoa L, Adolphs R. 2010. Emotion processing and the amygdala: from a “low road” to 835 \n“many roads” of evaluating biological significance. Nature Reviews Neuroscience 836 \n2010 11:11 11:773–782. doi:10.1038/nrn2920 837 \nPhelps EA, LeDoux JE. 2005. Contributions of the amygdala to emotion processing: 838 \nFrom animal models to human behavior. Neuron 48:175–187. 839 \ndoi:10.1016/J.NEURON.2005.09.025/ASSET/31E14E48-ACAC-4132-ACD2-840 \n31705A61F19C/MAIN.ASSETS/GR4.JPG 841 \nRafal RD, Koller K, Bultitude JH, Mullins P, Ward R, Mitchell AS, Bell AH. 2015. 842 \nConnectivity between the superior colliculus and the amygdala in humans and 843 \nmacaque monkeys: virtual dissection with probabilistic DTI tractography. J 844 \nNeurophysiol 114:1947–1962. doi:10.1152/JN.01016.2014 845 \nRamanathan KR, Jin J, Giustino TF, Payne MR, Maren S. 2018. Prefrontal projections 846 \nto the thalamic nucleus reuniens mediate fear extinction. Nature Communications 847 \n2018 9:1 9:1–12. doi:10.1038/s41467-018-06970-z 848 \nRamanathan KR, Maren S. 2019. Nucleus reuniens mediates the extinction of 849 \ncontextual fear conditioning. Behavioural Brain Research 374:112114. 850 \ndoi:10.1016/J.BBR.2019.112114 851 \nRatigan HC, Krishnan S, Smith S, Sheffield MEJ. 2023. A thalamic-hippocampal CA1 852 \nsignal for contextual fear memory suppression, extinction, and discrimination. 853 \nNature Communications 2023 14:1 14:1–17. doi:10.1038/s41467-023-42429-6 854 \nRolls ET, Huang CC, Lin CP, Feng J, Joliot M. 2020. Automated anatomical labelling 855 \natlas 3. Neuroimage 206. doi:10.1016/J.NEUROIMAGE.2019.116189 856 \nSaalmann YB, Kastner S. 2011. Cognitive and Perceptual Functions of the Visual 857 \nThalamus. Neuron 71:209. doi:10.1016/J.NEURON.2011.06.027 858 \nSaalmann YB, Kastner S. 2009. Gain control in the visual thalamus during perception 859 \nand cognition. Curr Opin Neurobiol 19:408. doi:10.1016/J.CONB.2009.05.007 860 \nSalay LD, Huberman AD. 2021. Divergent outputs of the ventral lateral geniculate 861 \nnucleus mediate visually evoked defensive behaviors. Cell Rep 37. 862 \ndoi:10.1016/j.celrep.2021.109792 863 \nSherman SM. 2016. Thalamus plays a central role in ongoing cortical functioning. 864 \nNature Neuroscience 2016 19:4 19:533–541. doi:10.1038/nn.4269 865 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 33 \nSherman SM. 2007. The thalamus is more than just a relay. Curr Opin Neurobiol 866 \n17:417–422. doi:10.1016/J.CONB.2007.07.003 867 \nSherman SMurray, Guillery RW. 2006. Exploring the thalamus and its role in cortical 868 \nfunction 484. 869 \nShi C, Davis M. 2001. Visual pathways involved in fear conditioning measured with fear-870 \npotentiated startle: behavioral and anatomic studies. J Neurosci 21:9844–9855. 871 \ndoi:10.1523/JNEUROSCI.21-24-09844.2001 872 \nSilverstein DN, Ingvar M. 2015. A multi-pathway hypothesis for human visual fear 873 \nsignaling. Front Syst Neurosci 9:137726. doi:10.3389/FNSYS.2015.00101/PDF 874 \nSteinmetz AB, Buss EW, Freeman JH. 2013. Inactivation of the ventral lateral geniculate 875 \nand nucleus of the optic tract impairs retention of visual eyeblink conditioning. 876 \nBehavioral Neuroscience 127:690–693. doi:10.1037/A0033729 877 \nTakakuwa N, Isa K, Onoe H, Takahashi J, Isa T. 2021. Contribution of the Pulvinar and 878 \nLateral Geniculate Nucleus to the Control of Visually Guided Saccades in Blindsight 879 \nMonkeys. Journal of Neuroscience 41:1755–1768. doi:10.1523/JNEUROSCI.2293-880 \n20.2020 881 \nTotty MS, Tuna T, Ramanathan KR, Jin J, Peters SE, Maren S. 2023. Thalamic nucleus 882 \nreuniens coordinates prefrontal-hippocampal synchrony to suppress extinguished 883 \nfear. Nature Communications 2023 14:1 14:1–12. doi:10.1038/s41467-023-42315-1 884 \nTovote P, Fadok JP, Lüthi A. 2015. Neuronal circuits for fear and anxiety. Nature 885 \nReviews Neuroscience 2015 16:6 16:317–331. doi:10.1038/nrn3945 886 \nVuilleumier P, Armony JL, Driver J, Dolan RJ. 2003. Distinct spatial frequency 887 \nsensitivities for processing faces and emotional expressions. Nature Neuroscience 888 \n2003 6:6 6:624–631. doi:10.1038/nn1057 889 \nWard R, Danziger S, Bamford S. 2005. Response to visual threat following damage to 890 \nthe pulvinar. Curr Biol 15:571–573. doi:10.1016/J.CUB.2005.01.056 891 \nWei P, Liu N, Zhang Z, Liu X, Tang Y, He X, Wu B, Zhou Z, Liu Y, Li J, Zhang Y, Zhou X, 892 \nXu L, Chen L, Bi G, Hu X, Xu F, Wang L. 2015. Processing of visually evoked 893 \ninnate fear by a non-canonical thalamic pathway. Nature Communications 2015 6:1 894 \n6:1–13. doi:10.1038/ncomms7756 895 \nWen Z, Pace-Schott EF, Lazar SW, Rosén J, Åhs F, Phelps EA, LeDoux JE, Milad MR. 896 \n2024. Distributed neural representations of conditioned threat in the human brain. 897 \nNature Communications 2024 15:1 15:1–14. doi:10.1038/s41467-024-46508-0 898 \nWen Z, Raio CM, Pace-Schott EF, Lazar SW, LeDoux JE, Phelps EA, Milad MR. 2022. 899 \nTemporally and anatomically specific contributions of the human amygdala to threat 900 \nand safety learning. Proc Natl Acad Sci U S A 119:e2204066119. 901 \ndoi:10.1073/PNAS.2204066119/SUPPL_FILE/PNAS.2204066119.SAPP.PDF 902 \nWhitfield-Gabrieli S, Nieto-Castanon A. 2012. Conn: a functional connectivity toolbox for 903 \ncorrelated and anticorrelated brain networks. Brain Connect 2:125–141. 904 \ndoi:10.1089/BRAIN.2012.0073 905 \nWhyte CJ, Redinbaugh MJ, Shine JM, Saalmann YB. 2024. Thalamic contributions to 906 \nthe state and contents of consciousness. Neuron 112:1611–1625. 907 \ndoi:10.1016/J.NEURON.2024.04.019/ASSET/AA96B159-0E29-4D7D-B3C8-908 \nE302EA442A73/MAIN.ASSETS/GR2.JPG 909 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 34 \nWolff M, Halassa MM. 2024. The mediodorsal thalamus in executive control. Neuron 910 \n112:893–908. doi:10.1016/J.NEURON.2024.01.002/ASSET/8EBBDE5D-737B-911 \n479F-966E-07AE8F092DB6/MAIN.ASSETS/GR3.JPG 912 \nWurtz RH, McAlonan K, Cavanaugh J, Berman RA. 2011. Thalamic pathways for active 913 \nvision. Trends Cogn Sci 15:177–184. doi:10.1016/J.TICS.2011.02.004 914 \nYarkoni T, Poldrack RA, Nichols TE, Van Essen DC, Wager TD. 2011. Large-scale 915 \nautomated synthesis of human functional neuroimaging data. Nature Methods 2011 916 \n8:8 8:665–670. doi:10.1038/nmeth.1635 917 \nZheng C, Huang Y , Bo B, Wei L, Liang Z, Wang Z. 2020. Projection from the Anterior 918 \nCingulate Cortex to the Lateral Part of Mediodorsal Thalamus Modulates Vicarious 919 \nFreezing Behavior. Neurosci Bull 36:217–229. doi:10.1007/S12264-019-00427-920 \nZ/METRICS 921 \n  922 \n 923 \n 924 \n 925 \n 926 \n 927 \n 928 \n 929 \n 930 \n 931 \n 932 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 35 \nFigure captions  933 \nFig 1. Neural representation of associative threat learning in pulvinar divisions, MD 934 \nand LGN.  935 \na. Human f ear conditioning paradigm in the fMRI. b-d. Means ± SE of activation in 936 \nresponse to CS+ vs. CS - at both block-wise and trial -wise levels within  the pulvinar 937 \ndivisions (b), MD (c), and LGN (d). e. ROI-to-ROI connectivity . Right: the red line 938 \nrepresents significant positive connectivity to CS+ vs. CS - (PFDR < 0.05), while the gray 939 \nlines indicate no n-significant connectivity.  Thalamic nuclei that showed no significant 940 \nconnectivity with other regions (i.e., amygdala, hippocampus, etc.) were omitted from the 941 \nvisualization. Left: boxplots and kernel density estimates illustrate the distribution of 942 \nconnectivity values in response to CS+ and CS-.  943 \nSE: Standard error.  944 \nMD: Mediodorsal thalamus.  945 \nLGN: Lateral geniculate nucleus.  946 \nCS+ vs. CS-: Conditioned stimulus predicting shock vs. no shock. 947 \nROI: Region of interest.  948 \n*p<0.05, **p<0.01, ***p<0.001 949 \nDisplay items in panel a were created using BioRender (BioRender.com). 950 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 36 \nFig 2. Quantifying the relationships between the anterior pulvinar and MD during 951 \nconditioning.  952 \na. Hierarchical regression models of trial-wise relationships between the anterior pulvinar 953 \nand MD activations while controlling for a potential effect of anatomical proximity. The 954 \neffect of anatomical proximity  was controlled  by progressively adding other pulvinar 955 \ndivisions as controls.  b. Results of a hierarchical model  that included the LGN as an 956 \nalternative control, distinct from  the pulvinar. c.  Comparison of activation levels in the 957 \nanterior pulvinar and MD at both block-wise and trial-wise levels (means ± SE).  958 \n*p<0.05, **p<0.01, ***p<0.001 959 \nMD: Mediodorsal thalamus.  960 \nLGN: Lateral geniculate nucleus.  961 \nSE: Standard error.  962 \n 963 \n 964 \n 965 \n 966 \n 967 \n 968 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 37 \nFig 3. A d ata-driven approach to understanding  the functional relationships 969 \nbetween pulvinar divisions during conditioning 970 \na. Means ± SE of a ctivation differences between pulvinar divisions. b. Network analysis 971 \nreveals that the medial pulvinar serves as a central hub, mediating interactions among 972 \npulvinar divisions and exhibiting increased centrality measures. c. Schematic 973 \nvisualization of activation patterns we observed in pulvinar divisions. d. Previous studies 974 \nsuggest that the inferior and lateral pulvinar are involved in processing basic visual 975 \ninformation, while the medial pulvinar is associated with higher -level functions, including 976 \nworking memory. e. Based on b-d, we hypothesize that the medial pulvinar mediates the 977 \nrelationships with other divisions. f. Mediation analysis supports our hypothesis (panel e). 978 \ng. Validation of the mediation model on an additional independent sample. Dashed paths 979 \nin panels f and g represent statistically unstable paths, while the continuous paths 980 \nindicate stable paths.  981 \nSE: Standard error.  982 \nSC: Superior colliculus.  983 \n*p<0.05, **p<0.01, ***p<0.001 984 \nDisplay items in panels d and e were created using BioRender (BioRender.com). 985 \n 986 \n 987 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 38 \n 988 \nFig 4. Neural representation of extinction learning in pulvinar divisions, MD and 989 \nLGN.  990 \na. Human extinction learning paradigm in the fMRI. b-d. Means ± SE of activation in 991 \nresponse to extinguished CS+ vs. CS- at both block-wise and trial-wise levels within the 992 \npulvinar divisions (b), MD (c), and LGN (d). e. ROI-to-ROI connectivity. Right: the red line 993 \nrepresents significant positive connectivity to extinguished CS+ vs. CS - (PFDR < 0.05), 994 \nwhile the gray lines indicate no significant connectivity. Thalamic nuclei that showed non-995 \nsignificant connectivity with other regions (i.e., amygdala, hippocampus , etc. ) were 996 \nomitted from the visualization. Left: boxplots and kernel density estimates illustrate the 997 \ndistribution of connectivity values in response to extinguished CS+ and CS-.  998 \nSE: Standard error.  999 \nMD: Mediodorsal thalamus.  1000 \nLGN: Lateral geniculate nucleus.  1001 \nExtinguished CS+ vs. CS-: A conditioned stimulus that no longer predicts shock vs. a 1002 \nstimulus that was never paired with shock. 1003 \nROI: Region of interest.  1004 \n*p<0.05, **p<0.01, ***p<0.001 1005 \nDisplay items in panel a were created using BioRender (BioRender.com). 1006 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 39 \n 1007 \nFig 5. Neural representation of extinction recall in pulvinar divisions, MD and LGN.  1008 \na. Human extinction recall paradigm in the fMRI conducted within safe contextual cues. 1009 \nb-d. Means ± SE of a ctivation in response to extinguished CS+ vs. CS - at both block -1010 \nwise and trial-wise levels within the pulvinar divisions ( b), MD (c), and LGN (d). e. ROI-1011 \nto-ROI connectivity. Right: t he red line represents significant positive connectivity to 1012 \nextinguished CS+ vs. CS - (PFDR < 0.05), while the gray lines indicate no n-significant 1013 \nconnectivity. Thalamic nuclei that showed no significant connectivity with other regions 1014 \n(i.e., amygdala, hippocampus, etc.) were omitted from the visualization. Left: boxplots and 1015 \nkernel density estimates illustrate the distribution of connectivity values in response to 1016 \nextinguished CS+ and CS-.  1017 \nSE: Standard error.  1018 \nMD: Mediodorsal thalamus.  1019 \nLGN: Lateral geniculate nucleus.  1020 \nExtinguished CS+ vs. CS -: A conditioned stimulus that no longer predicts shock vs. a 1021 \nstimulus that was never paired with shock. 1022 \nROI: Region of interest.  1023 \n*p<0.05, **p<0.01, ***p<0.001 1024 \nDisplay items in panel a were created using BioRender (BioRender.com). 1025 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 40 \n 1026 \nFig 6. Neural representation of threat renewal in pulvinar divisions, MD and LGN.  1027 \na. Human threat renewal paradigm in the fMRI conducted within threat contextual cues in 1028 \nthe original context where fear conditioning occurred . b-d. Means ± SE of activation in 1029 \nresponse to extinguished CS+ vs. CS- at both block-wise and trial-wise levels within the 1030 \npulvinar divisions (b), MD (c), and LGN (d). e. ROI-to-ROI connectivity. Right: the red line 1031 \nrepresents significant positive connectivity to extinguished CS+ vs. CS - (PFDR < 0.05), 1032 \nwhile the gray lines indicate non-significant connectivity. Thalamic nuclei that showed no 1033 \nsignificant connectivity with other regions (i.e., amygdala, hippocampus , etc. ) were 1034 \nomitted from the visualization. Left: boxplots and kernel density estimates illustrate the 1035 \ndistribution of connectivity values in response to extinguished CS+ and CS-.  1036 \nSE: Standard error.  1037 \nMD: Mediodorsal thalamus.  1038 \nLGN: Lateral geniculate nucleus.  1039 \nExtinguished CS+ vs. CS -: A conditioned stimulus that no longer predicts shock vs. a 1040 \nstimulus that was never paired with shock. 1041 \nROI: Region of interest.  1042 \n*p<0.05, **p<0.01, ***p<0.001 1043 \nDisplay items in panel a were created using BioRender (BioRender.com). 1044 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 41 \n 1045 \nFig 7. Neurobehavioral models of thalamic involvement in associative threat 1046 \nlearning and memory. 1047 \na. Schematic illustration of thalamic circuitry during the acquisition of associative threat 1048 \nlearning, highlighting interactions between pulvinar divisions, MD, and LGN with key brain 1049 \nregions involved in fear expression.  1050 \nb. A thalamic-dependent “toggle switch” regulates the retrieval of safety vs. threat-related 1051 \nmemory. The MD -dACC connectivity  modulates the interaction between dACC and 1052 \nvmPFC, promoting vmPFC dominance during extinction recall. In contrast, the anterior 1053 \npulvinar-vmPFC connectivity promotes dACC dominance, enhancing the expression of 1054 \nthreat memory during threat renewal.  1055 \nMD: Mediodorsal thalamus.  1056 \nLGN: Lateral geniculate nucleus.  1057 \nvmPFC = Ventromedial prefrontal cortex. 1058 \ndACC = Dorsal anterior cingulate cortex. 1059 \nV1, V2, and V4:  Primary, secondary, and fourth visual areas.  1060 \nTEO, TE: Temporal cortex regions.  1061 \nHypo.: Hypothalamus.  1062 \nHippo.: Hippocampus.  1063 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\n 42 \n 1064 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nBlocks comparison Trial-by-trial comparison\n**\n***\n***\n* ***\n**\n*\n*\n*****\n**\n***\n***\n***\n***\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n***\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\nMediodorsal thalamus\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n***\n***\nAnterior \nMedial\nInferior\nLateral\nPulvinar divisions\n \nLateral geniculate\nnucleus\nBlocks comparison Trial-by-trial comparison\nBlocks comparison Trial-by-trial comparison\nb c\nd\na\nAnterior\npulvinar\nAmygdala\ndACC\nvmPFC\nsgACC\nHippocampus\ne Connectivity\nAnterior pulvinar vmPFC Amygdala Hippocampus\n* * *\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nb\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\n*\n0.0\n0.1\n0.2\n0.3BOLD signal\nAnterior \nPulvinar MD\n***\nBlock comparison Trial-by-trial comparisons\nc\nAnterior pulvinar\nMD\nr = 0.66*** \nTrial 2\nAnterior pulvinar\nMD\nr = 0.66***\nRaw correlation \nHierarchical regression \nR2 change\nStandard 𝛃 estimations\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n1x104 bootstraps 𝛃 estimations\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\nTrial 1\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n1 2 3 4\nmodels\n1 2 3 4\nmodels\n44.1%\n1 2 3 4\nmodels\n1 2 3 4\nmodels\n43%\nAre the activation levels significantly different between anterior pulvinar and MD?\n*** *** *** \n*** \n*** ** \n*** *** *** \n*** \na\nR2 change = 0.006\nR2 change = 0.01\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n1x104 bootstraps\n 𝛃 estimations\nConservative control: Other pulvinar divisions Alternative control: LGN\nInferior pulvinar\nAnterior pulvinar Lateral pulvinar\nMedial pulvinar\nLGN\n*** \n*** \n*** \n*** \n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\nTrial 3\nAnterior pulvinar\nMD\nr = 0.77***\nAnterior pulvinar\nMD\nTrial 4\nr = 0.78***\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n1 2 3 4\nmodels\n1 2 3 4\nmodels\n1 2 3 4\nmodels\n1 2 3 4\nmodels\n59.4%\n60.7%\n*** \nR2 change = 0.03\nR2 change = 0.001\n-0.1\n0\n0.1\n0.2\n0.3\n0.4\n0.5\n0.6\n0.7\n0.8\n*** \n*** \n*** *** *** \n*** *** *** *** \n* * \n*** \n*** *** \n*** \n*** \n*** \n*** \n*** *** * \n** \n*** \n*** \n*** \n* \n* \n** \n*** *** \n*** \n*** \nIs the observed similarity in activation between the anterior pulvinar \nand MD statistically significant, or could it be attributed to \nanatomical proximity?\n1 2 3 4 1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nAnterior\n pulvinar MD\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nA = anterior pulvinar\nB = medial pulvinar\nC = inferior pulvinar\nD = lateral pulvinar\n \nBOLD signal = CS+ minus CS-\n \ne\nB-C\nB-A\nD-A\nB-D\nC-D\nC-A\nSample\nBootstrap mean \nEdges\nA\nB\nC\nD\nC\nB\nA\nB\nC\nD\n1x104 Bootstraps\n \nCentrality differences\nbased on 1x104 Bootstraps\n \nB\nC\nStrength Closeness\nB\nC\nD\nA\nB\nC\nBetweenness\nNot stable\nStable\nba\nActivation\nCentrality measures\nZ-scores \nTrial 1 Trial 2 Trial 1 Trial 2\nCS+\nCS-\nAnterior\n pulvinar\nOther \nsub-pulvinars\nSchematic representation\nOur findings of the activation pattern in \nconditioning\nNetwork plots\nAnterior pulvinar may integrate \ninformation after initial processing in \nother sub-pulvinars. \nMedial pulvinar may act as a relay or convergence point \nbetween the sensory-oriented inferior and lateral pulvinars \nand the integrative anterior pulvinar.\nUnderstudied \npaths\nLateral \npulvinar\nMedial \npulvinar\nInferior \npulvinar\nAnterior \npulvinar\n…\ndc Previous studies showed:\nNetwork stability\nA B C D\n0.0\n0.1\n0.2\n0.3BOLD signal\n-2 -1 0 1 2\nA\nB\nC\nD\n-0.4 -0.2 0 0.2 0.4 0.6 0.8\n| | | | | | |\ntest\nHypothesis\nMediation model using K-fold cross-validation techniques f\nLateral \npulvinar\nMedial \npulvinar\nInferior \npulvinar\nAnterior \npulvinar\n0.49 0.25\n0.48 -0.12\n0.73\nTraining Test Training Test Training Test\n3 folds\nRandom allocation into 3-folds\nSample\nN = 293\nSub-samples\neach n ~ 91- 106\nPath coefficient \nOverall sample\nAggregated paths coefficients\n-0.4\n-0.2\n0.0\n0.2\n0.4\n0.6\n-0.4\n-0.2\n0.0\n0.2\n0.4\n0.6\n-0.4\n-0.2\n0.0\n0.2\n0.4\n0.6\nR2 \nSample\nLateral \npulvinar\nMedial \npulvinar\nInferior \npulvinar\nAnterior \npulvinar\n0.42 0.25\n0.49 -0.05\n0.71\nIndependent cohortg\nN = 114\nR2 \n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nBlocks comparison Trial-by-trial comparison\nMediodorsal thalamus\nAnterior \nMedial\nInferior\nLateral\nPulvinar divisions\n \nLateral geniculate\nnucleus\nBlocks comparison\nBlocks comparison Trial-by-trial comparison\nb c\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\n**\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\n*\n*\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n*\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n** \nTrial-by-trial comparison\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n** \nExtinguished \na\nd\nLGN\nHippocampus\nAmygdala\ndACC\nvmPFC\nsgACC\ne Connectivity\nLGN sgACC\n*\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nMediodorsal thalamus\nAnterior \nMedial\nInferior\nLateral\nPulvinar divisions\n \nLateral geniculate\nnucleus\nBlocks comparison\nBlocks comparison Trial-by-trial comparison\nb c\nExtinguished \nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\n*\n*\nTrial-by-trial comparisonBlocks comparison\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n**\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n***\nTrial-by-trial comparison\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n*\n** \nd\na\nvmPFC\nMediodorsal \nthalamus\nAmygdala\ndACC\nsgACC\nHippocampus\ne Connectivity\nMediodorsal thalamus dACC\n*\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nMediodorsal thalamus\nAnterior \nMedial\nInferior\nLateral\nPulvinar divisions\n \nLateral geniculate\nnucleus\nBlocks comparison\nBlocks comparison\nb c\nExtinguished \nBlocks comparison\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n**\nTrial-by-trial comparison\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n*\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n**\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\n** *\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n***\nTrial-by-trial comparison\n1 2 3 4\n-0.2\n0.0\n0.2\n0.4\n0.6\n0.8BOLD signal\nTrials\n***\nCS+ CS-\n0.0\n0.1\n0.2\n0.3\n0.4BOLD signal\n*\nTrial-by-trial comparison\na\nd\nAmygdala\ndACC\nvmPFC\nsgACC\nAnterior\npulvinar\nHippocampus\ne Connectivity\nAnterior pulvinar vmPFC\n**\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint \n\nLateral \npulvinar\nMedial \npulvinar\nInferior \npulvinar\nAnterior \npulvinar\nSuperior \ncolliculus\nHippo. Amygdala\nvmPFC\nLGN V1 TEOV2 V4 TE\nMD\nPotential neural \ngate for \nconscious and \nsubconscious \nthreat learning \nCo-activation\nIntegrative simplified functional thalamic circuitry during the acquisition of associative threat learning \nMotor dACC\nBehavioral output \nCS\nSensory \ninputs\nAssociative threat learning \nEmotional output \nFight\nFlight\nFreeze\nFear\nHypo.\n+ Hormonal \nresponse\nExtinction recall - safe cues\nHippocampus\nMD\n Anterior \npulvinar\n MD Anterior \npulvinar\nAmygdala\nExcitationInhibition\nThreat memory expression Safe memory expression \nThreat renewal - threat cues\nDynamic model of safe vs. threat memory retrieval using contextual cues: \nA thalamic-dependent “toggle switch”\na\nb\nHippocampus\n.CC-BY 4.0 International licenseavailable under a \nwas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprint (whichthis version posted July 14, 2025. ; https://doi.org/10.1101/2025.07.09.663823doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}