Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences Jan Haakera,b,1,2 , Lorenzo Diaz-Mataixc,d,1,2 , Gemma Guillazo-Blanche , Sara A. Starkc , Lea Kernb , Joseph E. LeDouxc,d,f , and Andreas Olssonb a Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, 20246 Hamburg, Germany; b Division of Psychology, Department of Clinical Neuroscience, Karolinska Institutet, 171 77 Stockholm, Sweden; c Center for Neural Science, New York University, New York, NY 10003; d Emotional Brain Institute, New York University, New York, NY 10003; e Departament de Psicobiologia i Metodologia de les Ciències de la Salut, Institut de Neurociències, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain; and f Department of Psychology, New York University, New York, NY 10003 Edited by Huda Akil, University of Michigan, Ann Arbor, MI, and approved May 12, 2021 (received for review January 26, 2021) Information about dangers can spread effectively by observation ment USs were presented (12). While such reinstatement of of others’ threat responses. Yet, it is unclear if such observational threat responses after firsthand aversive (US) experiences is threat information interacts with associative memories that are a central mechanism for adaptive and maladaptive behavior shaped by the individual’s direct, firsthand experiences. Here, we toward threats (13, 14), it is unclear if information about others’ show in humans and rats that the mere observation of a con- aversive experiences can disinhibit threat responses and rein- specific’s threat reactions reinstates previously learned and extin- state firsthand acquired threat memories. This knowledge would guished threat responses in the observer. In two experiments, reveal how social information interacts with firsthand learning of human participants displayed elevated physiological responses threats and could thereby provide a mechanistic perspective for to threat-conditioned cues after observational reinstatement in a overlapping processes that underlie observational and firsthand PSYCHOLOGICAL AND COGNITIVE SCIENCES context-specific manner. The elevation of physiological responses learning. (arousal) was further specific to the context that was observed as We examined whether observation of others’ reactions to dangerous. An analogous experiment in rats provided converg- threats recovers directly, firsthand acquired and extinguished ing results by demonstrating reinstatement of defensive behavior threat responses (using Pavlovian threat conditioning) in rats after observing another rat’s threat reactions. Taken together, our and humans and thereby reinstates conditioned threat memo- findings provide cross-species evidence that observation of oth- ries in both species. We hypothesized that observation of others’ ers’ threat reactions can recover associations previously shaped by behavior toward a US reinstates CRs in humans and rats. direct, firsthand aversive experiences. Our study offers a perspec- tive on how retrieval of threat memories draws from associative Results mechanisms that might underlie both observations of others’ and Observation of Others’ Reactions to the US Reinstates Conditioned firsthand experiences. Threat Responses in Humans. The first experiment was designed to establish that CRs return after extinction by mere observa- reinstatement | vicarious learning | social learning | threat conditioning tion of another individual who reacts to a US (i.e., observational Significance S ocial transmission of information about dangers in the envi- ronment is central for the adaption of defensive behaviors (1–3). As such, observing a conspecific’s threat responses to a Social transmission of threat information by observation is effective in humans and other animals. However, it is stimulus or context commonly results in aversive learning and unknown if such observation of others’ reacting to threats avoidance, saving the observer from costly learning from first- can retrieve memories that have been previously learned hand experienced trial and error (4–6). Yet, it is still unanswered through direct, firsthand aversive experiences. Here, we show if such social threat learning interacts with previous memo- concordantly in humans and rats that observing a conspe- ries resulting from firsthand experiences. In other words, can cific’s reactions to a threat is sufficient to recover associative social information recover responses that have been learned by memories of direct, firsthand aversive experiences, measured firsthand experiences? as conditioned threat responses (physiological responses and To address this open question, we tested in rats and humans defensive behavior) in the observer. The reinstatement of if watching another individual’s threat reactions leads to recov- threat responses by observation of others is specific to the ery (or reinstatement) of conditioned responses to a threat context that is observed as being dangerous. Our findings pro- cue that was previously acquired through firsthand aversive vide cross-species evidence that observation of others’ threat experience (i.e., Pavlovian threat conditioning). Our obser- reactions can recover associative memories of direct, firsthand vational reinstatement procedure was inspired by established aversive experiences. paradigms using firsthand aversive experience to reinstate con- ditioned defensive responses (CRs) in humans and other Author contributions: J.H., L.D.-M., J.E.L., and A.O. designed research; J.H., L.D.-M., animals (7–11). G.G.-B., S.A.S., and L.K. performed research; J.H. and L.D.-M. analyzed data; and J.H., In the reinstatement paradigm, individuals first acquire an L.D.-M., G.G.-B., J.E.L., and A.O. wrote the paper with comments from all authors.y association between a cue (conditioned stimulus [CS]) that is The authors declare no competing interest.y predictive of a firsthand aversive experience (unconditioned This article is a PNAS Direct Submission.y stimulus [US]), which results in CRs toward the CS. When the This open access article is distributed under Creative Commons Attribution-NonCommercial- CS is no longer followed by the US (and hence, no longer predic- NoDerivatives License 4.0 (CC BY-NC-ND).y tive of the US), CRs attenuate or extinguish slowly. However, the 1 J.H. and L.D.-M. contributed equally to this work.y initial CS–US association is not thought to be erased but rather, 2 To whom correspondence may be addressed. Email: j.haaker@uke.de or inhibited. This inhibition can be abolished by presentation of the ldmataix@gmail.com.y US alone (reinstatement), which reinstates the CR to the CS. This article contains supporting information online at https://www.pnas.org/lookup/suppl/ Importantly, the reinstatement of the initial CS–US association doi:10.1073/pnas.2101290118/-/DCSupplemental.y is critically dependent on the context in which the reinstate- Published July 23, 2021. PNAS 2021 Vol. 118 No. 30 e2101290118 https://doi.org/10.1073/pnas.2101290118 | 1 of 8 reinstatement of CRs). We included two groups that under- Appendix, Table S4). Hence, these analyses supported a rein- went identical acquisition and extinction training that differed statement (or increase of CRs after extinction) as a function of only with respect to the demonstrator receiving USs or no USs, observing the demonstrator reacting to the US. A post hoc test respectively (the design is in Fig. 1A). further confirmed our hypothesis that SCRs in the experimen- Participants (N = 35) in both groups acquired higher threat tal group during the reinstatement test were enhanced to the responses toward a CS+ that was followed by an aversive US CS+ and the CS– when compared with the control group [one- (aversive electrical stimulation) as compared with a control stim- sided t test CS+: t(33) = −2.513, pcorr = 0.018, Cohen’s d = ulus (CS–; not followed by a US) measured as skin conductance –0.850, mean difference = –0.259 ± 0.10 (SE), upper 95% CI: responses [SCRs; main effect of CS type, F(1,33) = 5.71, P = –0.085; CS–: t(33) = −2.90, pcorr = 0.009, Cohen’s d = –0.982, 0.023; ηp2 = 0.148, CS+ > CS–: t(33) = 2.39, pcorr = 0.022; mean difference = –0.157 ± 0.05 (SE), upper 95% CI: –0.065] Cohen’s d = 0.405; mean difference = 0.130 ± 0.054 (SD), (Fig. 1C and SI Appendix, Table S4; SI Appendix, Table S5 has 95% CI: 0.020/0.249] with no main effect or interaction with enhanced SCRs after observational reinstatement when con- the factor group (P > 0.3) (SI Appendix, Fig. S1 shows details sidering the whole reinstatement test phase). This effect was and blockwise SCRs). Subsequently, during extinction training, concordant with an enhanced expectancy for the US after obser- the CS+ was no longer followed by a US, leading to attenu- vational reinstatement as compared with the no-reinstatement ated SCRs, albeit remaining elevated to the CS+ as compared control group [one-sided t test CS+: t(32) = 2.141, pcorr = 0.040; with the CS– [main effect of CS type: F(1,33) = 9.48, P = CS–: t(32) = 0.272, pcorr > 0.7], assessed by a postexperimental 0.004; ηp2 = 0.223; CS+ > CS−: t(33) = 3.06, pcorr = 0.004; questionnaire. Cohen’s d = 0.517; mean difference = 0.089 ± 0.029 (SD), In sum, the results of our first experiment suggest that observ- 95% CI: 0.030/0.149], again without a main effect or interac- ing another individual receiving a US reinstates conditioned tion with the factor group (P > 0.16) (SI Appendix, Table S2 has threat responses that have been previously acquired by direct US details). experiences. Importantly, participants who observed a demonstrator model reacting to the US after extinction (experimental group) showed Observational Reinstatement Is Contingent on Contextual Infor- enhanced SCRs as compared with participants in the control mation. In the second experiment (N = 21), we tested if group who observed a calm demonstrator [late extinction (sec- observational reinstatement displayed a key characteristic of ond block) to the reinstatement test (first block), group by direct reinstatement, namely contextual dependence. Previous phase interaction: F(1,33) = 4.90, P = 0.034; ηp2 = 0.129 and research has established that enhancement of CRs is specific main effect of group: F(1,33) = 9.136, P = 0.005; ηp2 = 0.22] to the context in which the participants received firsthand (Fig. 1B and SI Appendix, Table S3). This interaction effect reinstatement USs [humans (15); animals (7)]. Indeed, rein- was indicative of the hypothesized increase in SCRs from late statement USs are thought to render a context dangerous, extinction to reinstatement test in the experimental group as which gates reinstatement of CRs at a later test. Accordingly, compared with the control group [one-sided t test: t(33) = we examined if observation of reinstatement within a specific −2.213, pcorr = 0.017, Cohen’s d = –0.748, mean difference = context (here, red illuminated room) leads to enhancement –0.156 ± 0.07 (SE), upper 95% CI: –0.037] (Fig. 1C and SI of CRs within the (red) context in which the demonstrator A Observational Reinstatement Group Observational CS+ C CS- CS+ CS- Control Group CS+ CS- Acquisition Reinstatement procedure Retention Test Extinction (RI-test) B Group * Phase: p=0.034 C 0.7 CS+ CS- CS+ CS- SCR [lg max corrected] CS+ CS+ p=0.018 0.5 CS- (N=17) CS- (N=18) 1.5 p=0.009 0.3 1.0 0 0.5 0.5 0.1 0 0 0 ACQ EXT EXT TEST ACQ EXT EXT TEST T T T t t t T t es es es es EX EX EX EX -T -T -T -T RI RI RI RI Control Observational Control Observational Group Reinstatement Group Reinstatement Fig. 1. Observational reinstatement of firsthand acquired CRs. (A) Overview of the study design in experiment 1. Yellow lightning bolts denote firsthand aversive electrical stimulation to the observing participant. The red dotted lightning bolt denotes aversive electrical stimulation to the demonstrator, serving as an observational US for the observing participant. (B) Bar graphs representing the SCRs during acquisition, early (first block) and late extinction (second block), and the reinstatement test (first block) in the “no-reinstatement control” group (Left) and the observational reinstatement group (Right). SI Appendix, Fig. S1 shows blockwise SCRs during acquisition and the reinstatement test. Error bars represent the SEM. (C) Box plots of individual SCRs during late extinction (second block) and the reinstatement test (first block) for each of the CS+ in the control group (left four bars) and the observational reinstatement group (right four bars). P values indicate interaction in ANOVA (rmANOVA in B) or post hoc t test (one sided in C, corrected for multiple comparisons using Bonferroni–Holm). ACQ = acquisition training; EXT = extinction training. 2 of 8 | PNAS Haaker et al. https://doi.org/10.1073/pnas.2101290118 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences reacted defensively toward the US. Thereby, we would pro- ited enhancement of CRs from late extinction (second block) vide insights into how social information can exert contextual to the reinstatement test in the reinstatement context B [one- gating of CRs. sided comparison: late extinction vs. context B: t(20) = −2.621, To address the context dependence of observational rein- pcorr = 0.025, Cohen’s d = –0.572, mean difference = –0.330 ± statement, another sample of participants underwent firsthand 0.13 (SE), upper 95% CI: –0.001; late extinction vs. context acquisition and extinction training in context A followed by A: t(20) = −1.702, pcorr = 0.052, Cohen’s d = –0.371, mean observation of the demonstrators’ reactions to the US in context difference = –0.152 ± 0.089 (SE), upper 95% CI: 0.081; con- B (reinstatement context) (Fig. 2A and Materials and Meth- text A vs. context B: t(20) = −2.371, pcorr = 0.028, Cohen’s ods). Participants were then tested in contexts A and B for d = –0.517, mean difference = –0.178 ± 0.075 (SE), upper 95% reinstatement of conditioned responses. Thus, context speci- CI: 0.018] (Fig. 2C and SI Appendix, Table S11 have details). ficity of reinstatement was tested by within-subject comparisons Hence, reinstatement of responses was evident in the reinstate- of responses when participants were placed in context A and ment context (i.e., the context in which the demonstrator was context B, respectively (cross-over, counterbalanced design). present during observational reinstatement). We found weak We expected context-specific reinstatement of CRs during the evidence for a stimulus-specific effect of observational rein- test, evident by enhanced CRs in context B (the context of statement in the rmANOVA [CS type × context interaction: the observational reinstatement) as compared with control con- F(2,40) = 2.953, P = 0.082; ηp2 = 0.129] that was characterized as text A. During acquisition training, participants acquired higher enhancement of responses for both the CS+ and the CS– in the responses toward the CS+ as compared with the CS–, mea- reinstatement context B [one-sided comparison: late extinction sured as SCR [CS type: F(1,20) = 7.987, P = 0.010; ηp2 = 0.285; vs. context B CS+: t(20) = −2.398, pcorr = 0.039, Cohen’s d = CS+ > CS–: t(20) = 2.826, pcorr = 0.010, Cohen’s d = 0.617, –0.523, mean difference = –0.228 ± 0.10 (SE), upper 95% CI: mean difference = 0.191 ± 0.07 (SE), 95% CI: 0.050/0.331] (SI –0.064; CS–: t(20) = 2.570, pcorr = 0.036, Cohen’s d = 0.561, Appendix, Fig. S4 shows blockwise SCRs, and SI Appendix, Table mean difference = 0.432 ± 0.17 (SE), upper 95% CI: –0.142] S9 has details). Subsequent extinction training led to decreas- (Fig. 2C and SI Appendix, Table S12). Such a generalized PSYCHOLOGICAL AND COGNITIVE SCIENCES ing responses [main effect of block: F(1,20) = 5.315, P = 0.032; responding after reinstatement (i.e., to the CS+ and the CS–) ηp2 = 0.210; Block 1 > Block 2: t(20) = 2.305, pcorr = 0.032, has been frequently reported in humans (8) and might actually be Cohen’s d = 0.503, mean difference = 0.081 ± 0.04 (SE), 95% a result of the strong contextual influence in this experiment that CI: 0.008/0.115] (SI Appendix, Table S10), with no support for affects all stimuli that are presented within the dangerous con- a difference between CSs (main effect of CS type: P > 0.19). text (8, 16). Control analyses, including the order of the contexts Importantly, when comparing the SCRs during late extinction in in reinstatement testing as well as accounting for individual dif- context A with responses during the reinstatement test in control ferences in both dispositional and current anxiety, still revealed context A and in reinstatement context B, analyses revealed a a contextual effect on observational reinstatement (SI Appendix, main effect of context [main effect of context: F(2,28.7) = 5.554, Tables S13–S15). In accordance with experiment 1, the postex- P = 0.016; ηp2 = 0.217; the factor context includes three levels: perimental ratings revealed a differential reinstatement effect extinction, context A, context B] (Fig. 2B and SI Appendix, Table for the CS+ in retrospectively assessed expectations of the US S11). In accordance with our hypothesis, participants exhib- (P < 0.001). A B C Fig. 2. Observational reinstatement (SCR) is context specific. (A) Overview of the study design in experiment 2. Yellow lightning bolts denote firsthand aversive electrical stimulation to the observing participant. The red dotted lightning bolt denotes aversive electrical stimulation to the demonstrator, serving as an observational US for the observing participant. (B) Bar graphs representing the SCRs during acquisition, early (first block) and late extinction (second block), and the reinstatement test in context A (white lamp) and context B (red lamp). P values (corrected for multiple comparisons by Bonferroni–Holm) indicate comparisons between late extinction and both contexts. SI Appendix, Fig. S4 shows blockwise SCRs during acquisition. Error bars represent the SEM. (C) Box plots of individual SCRs to both CSs during late extinction (second block) and the reinstatement test (first block) in the reinstatement context. ACQ = acquisition training; EXT = extinction training. Haaker et al. PNAS | 3 of 8 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences https://doi.org/10.1073/pnas.2101290118 In sum, our results from two studies in humans pro- −11.788 ± 1.12 (SE), 95% CI: −14.112/−9.464] (SI Appendix, vide evidence for contextual-dependent reinstatement of threat Table S20 has details). There was no difference in freezing responses after observation of a demonstrator reacting to a US, behavior between demonstrators and observers in the extinc- indicated by both psychophysiological measures (SCR) and US tion retention test (no interaction or main effect, P > 0.62) (SI expectancy. Interestingly, responses are returning in a context Appendix, Table S22). For the observational reinstatement pro- that is potentially dangerous for the demonstrator, but was never cedure on the next day (day 5), rats (previous cage-mate pairs) predictive for a US to the participants. were placed in two adjacent chambers, which were separated Next, we wanted to examine observational reinstatement in by plexiglass with holes that allowed for olfactory and auditory an experimental model where the synaptic substrates of first- exchange (rat pairs were habituated to this chamber without hand reinstatement have been established (17, 18). To this end, a US before acquisition training). One of the rats (demon- we tested for defensive responses measured as freezing behav- strator rat) received four unsignaled foot shocks, whereas the ior (which is commonly examined as a CR in rodents) after rat (observer rat) in the other chamber did not receive any firsthand and observational reinstatement procedures in rats foot shocks. Strikingly, we found that freezing behavior to the (experiment 3). We adapted the observational reinstatement extinguished CS+extinguished increased in both the demonstrator protocol to meet species-specific demands, which allowed us to rats that received firsthand US reinstatement and the observer examine if observational reinstatement is robust to such changes. rats that observed the reinstatement procedure [main effect of Our hypothesis was that rats, as humans, show observational reinstatement: F(1,21) = 4.802, P = 0.040; ηp2 = 0.186] (Fig. reinstatement of defensive responses that are comparable with 3B and SI Appendix, Table S21). Specifically, freezing behavior firsthand reinstatement. increased from the extinction test (day 4) to the reinstatement test (day 6) to the CS+extinguished and reached the same level of Observational Reinstatement of Conditioned Responses Is Evident the CS+nonextinguished [CS type × reinstatement: F(1,21) = 38.186, in Rats. In experiment 3, 23 (11 pairs) of male cage-mate rats P < 0.001; ηp2 = 0.642; CS+extinguished increase: t(21) = −6.110, individually underwent acquisition training, in which a flashing pcorr = 1.294e-6, mean difference = −6.366 ± 1.042, 95% CI: light and a tone were both predictive of a foot shock US (day −9.257/−3.474; CS+nonextinguished decrease: t(21) = 3.580, pcorr = 1). Twenty-four hours later (day 2), the rats underwent extinc- 0.003, mean difference = 3.580 ± 1.042, 95% CI: 0.688/6.471; tion training to one of the CS+ (CS+extinguished ) on 2 consecutive CS+extinguished vs. CS+nonextinguished during reinstatement test: days (Fig. 3A). Behavioral analyses of the retention test (day 4) t(21) = −1.649, pcorr = 0.107, mean difference = −1.843 ± 1.117, revealed successful acquisition and extinction training, indicated 95% CI: −4.937/1.251] (Fig. 3B and SI Appendix, Tables S21 and by longer duration of freezing behavior to the nonextinguished S22; SI Appendix, Tables S23 and S24 have separate post hoc CS+nonextinguished when compared with freezing to the extinguished tests in demonstrators and observers). These results show that CS+extinguished , respectively [CS type: F(1,21) = 112.196, P < observational reinstatement, which we found in humans in exper- 0.001; ηp2 = 0.842; CS+extinguished > CS+nonextinguished : t(20) = iments 1 and 2, is also evident in rats. Interestingly, we found no 10.548, pcorr < 0.001, Cohen’s d = −2.199, mean difference = statistical support for a difference in reinstatement of freezing A B C Fig. 3. Observational reinstatement in rats. (A) Overview of the study design in experiment 3 in rats that included acquisition of two CS+–US associations (using a light CS and a tone CS), one of which was extinguished (CS+extinguished), whereas the other was not (CS+nonextinguished). Yellow lightning bolts denote firsthand aversive electrical stimulation to all rats during acquisition. The red dotted lightning bolt denotes aversive electrical stimulation to the demonstrator rats, indicating an observational US to the observer rats. (B) Bar graphs illustrating an increase in freezing behavior to the CS+extinguished from the retention test before (day 4 extinction test; second block) to after observational reinstatement (day 6 RI test; first block). This increase was evident in both demonstrators (gray bars) and observers (blue bars) to the extinguished CSextinguished but not to the CSnonextinguished. There was no support for differences between demonstrators and observers with respect to freezing behavior in the RI test on day 6. (C) Box plots of individual freezing duration during the extinction test (second block) and the reinstatement test (first block) for each the CSextinguished in the observers (left two bars) and the demonstrators (right two bars). EXT = extinction retrieval test; RI-Test = reinstatement test. 4 of 8 | PNAS Haaker et al. https://doi.org/10.1073/pnas.2101290118 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences behavior between observers and demonstrators (all interactions: which was, however, never predictive for a US to the partici- P > 0.15) (SI Appendix, Table S21), although observers dis- pants. Hence, the observational return of cued threat responses played less freezing when compared with demonstrators across was transferred to a context, which has merely been observed as all phases [main effect of being observer or demonstrator: dangerous to another individual. Such a transfer of previously F(1,21) = 6.774, P = 0.017; ηp2 = 0.244; demonstrators vs. shaped firsthand association by contextual information that is observers: t(21) = 2.603, pcorr = 0.017, Cohen’s d = 0.543, mean derived from observation of others might be particularly adaptive difference = 2.052 ± 0.79, 95% CI: 0.412/3.692]. In other words, to generalize threat responses to novel dangerous situations. our results do not reveal any difference in CRs between rats that Previous research in humans showed that reinstatement of received a firsthand US vs. observed a cage-mate receiving the CRs is possible with exposure to a US that had not initially US. SI Appendix has an additional experiment to explore the been paired with the CS, so-called “cross-US reinstatement” expression of immediate early genes (c-fos protein) in rats within (22). However, there exist several differences between studies a similar protocol as in experiment 3. of such cross-US reinstatement and our results. First, cross-US reinstatement leads to expectancy for the US that was presented Discussion during reinstatement (22). In contrast, participants in our study Our findings provide cross-species evidence that observation still rated (in a postexperimental interview) high expectancy for of others’ threat reactions can recover associations previ- the firsthand US, which was used during acquisition. Second, ously shaped by direct, firsthand aversive experiences. Specifi- whereas cross-US reinstatement using different US types has cally, in three experiments studying both humans and rats, we not been found in rats (17, 18), our results provide evidence show that the observation of a conspecific exposed to a US for observational reinstatement in rats. Instead, we suggest that can recover—“reinstate”—previously extinguished conditioned observation of others’ responses to a US interacts with previ- responses (SCR in humans and freezing behavior in rats). Impor- ously made associations of firsthand US experiences, which in tantly, the recovery of threat responses indicates that observing turn, jointly gate recovery of threat responses. In support of such a conspecific’s US reaction overrides the previously learned a mechanism underlying observational reinstatement are pre- PSYCHOLOGICAL AND COGNITIVE SCIENCES inhibition of threat responses by firsthand experiences of the vious findings of reinstated threat responses through retrieval absence of the US (i.e., extinction training). Social transmis- of firsthand US associations (instead of a firsthand US experi- sion of threat information might therefore not only recover ence). In particular, reinstatement in rodents has been found associations from firsthand aversive experiences but also dimin- after presentation of a conditioned CS (23) or a conditioned con- ish safety information that we learned from our own firsthand text (24, 25). Similarly, the observation of US responses in our experience. Moreover, our results indicate that the recovery of experiments might also have signaled (or fostered retrieval of) threat responses through observation is context specific so that association of the firsthand US experience. In fact, responses to responses are displayed only in a situation that is potentially others’ expression of distress are often explained by associative dangerous for the observed individual. We furthermore provide learning mechanisms and thought to underlie association with concordant results across humans and rats using species-specific experiences of firsthand aversive events (26, 27). Our experi- adaptions of observational protocols and complementary mea- mental evidence that the observational reinstatement procedure surements of conditioned responses (i.e., psychophysiological leads to recovery of what is learned from firsthand experience responses and defensive behavior, respectively) (19). Our find- would speak in favor of a common associative learning mech- ings indicate that observational reinstatement of conditioned anism in observational and firsthand learning. While we have responses is robust to changes in the protocol (e.g., live interac- employed extinction learning protocols with similar numbers of tion vs. video clips, differential conditioning vs. nonextinguished trials (28, 29) and there is no support for a relationship between cues, immediate vs. delayed extinction). Note that the terms the amount of extinction trials and the strength of reinstate- “firsthand experience” and “observational experience” do not ment (8), future studies need to determine if prolonged (first- necessitate conscious experiences, in particular when examining hand or observational) extinction might influence observational peripheral arousal in humans and freezing behavior in animals. reinstatement. Our results provide experimental evidence for the return of Our findings and the suggested interaction between learned threat associations solely through social information. Thereby, association from firsthand experiences and processing of sig- our results extend current conditions for the return of threat nals from conspecifics are consistent with a previous finding in responses (8, 11, 12, 14, 20, 21) that include shifts of the con- rodents that revealed enhanced acquisition of CR after inter- text (i.e., renewal), passage of time (i.e., spontaneous recovery), acting with conspecifics that underwent acquisition training (30). CS–US pairings (rapid reacquisition), and direct experience of Our finding that observation of others’ responses to threats mod- the US (i.e., direct reinstatement). Moreover, we demonstrate ulates the expression of defensive behaviors also aligns with that observational reinstatement shares a defining characteris- research demonstrating enhanced recovery (here, contextual tic with direct reinstatement, namely contextual dependency. renewal) of CRs in an observer rat tested in the presence of a Indeed, the context where the observational reinstatement takes conspecific that exhibited defensive responses (31). Moreover, place (i.e., the context that surrounds the demonstrator) modu- our results are consistent with reports in humans that previ- lates reinstatement of conditioned responses, similar to contex- ous firsthand experiences modulate responses toward aversive tual specificity of direct reinstatement (7, 15) and other forms experience in others (32) and that observation of others’ dis- of threat recovery (12, 14, 21). Such strong contextual influ- tress biases choices that depend on firsthand experiences (33) ence might have been one reason for the increase in SCRs since direct and observational processes interact during decision to the CS– during the reinstatement test in experiment 2 (i.e., making (34). generalized reinstatement), which is a common reinstatement It should be stressed that our results cannot be explained by phenomenon in human research (8). We further found that pos- imitation or social facilitation (i.e., when responses of an indi- texperimental US expectancy was increased for the CS+ (and vidual are modulated by the passive presence of a conspecific) not the CS–) after reinstatement, which might be due to the because the demonstrator model is absent during the test. Nei- different sampling time points and the divergence in processes ther can our results be explained by stimulus enhancement of that underlie rating of US expectancy and psychophysiological responses to the CS (i.e., when observation of responses to an arousal. object facilitates acquisition of the observed response) because Interestingly, we found return of threat responses in the con- no CS was present during reinstatement (5). The experiments text in which the demonstrators’ threat reaction was observed, in humans used a relatively low number of extinction trials Haaker et al. PNAS | 5 of 8 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences https://doi.org/10.1073/pnas.2101290118 [as in previous experiments (28, 29)] that might have facili- associative learning mechanisms that gate threat responses in tated reinstatement, even though the number of trials during social contexts. extinction has not been found to impact the strength of the reinstatement (8, 35). It is theoretically possible that our rela- Materials and Methods tively low number of extinction trials in the human experiments Human Experiments. might have facilitated reinstatement. Arguing against this pos- Participants. Participants (no intake of prescription medication, no current sibility, however, is the finding in a recent review of the human or prior psychiatric disorders or neurological disorders) were recruited via reinstatement literature that a higher number of extinction, but advertisements at local universities in Stockholm as well as via a website. Thirty-nine participants were recruited in experiment 1, and 24 participants not conditioning, trials is overrepresented in studies report- were recruited in experiment 2. Participants were reimbursed with two cin- ing differential reinstatement (and not a reduced reinstatement ema tickets. Subjects were excluded for analyses if they could not report effect). Moreover, a study that statistically compared different the stimulus contingencies (Procedure). In experiment 1, this resulted in numbers of extinction trials during extinction found no impact exclusion of four subjects (control group = 1, experimental group = 3), result- of the amount of extinction on the degree of the reinstatement ing in a sample of 35 participants (19 females, age between 19 and 30 y, (measured as startle) (35). Furthermore, we used a more exten- mean = 23.8, SD = 2.9) in total. In experiment 2, this resulted in exclusion sive extinction protocol in the experiment in rats (2 d, each with of three subjects, resulting in a sample of 21 participants (16 female, age 15 CS presentations), which renders it rather unlikely that the between 18 and 28 y, mean = 22.2, SD = 2.9) in total. Exclusion of partici- effect of extinction trials had a strong influence on the observa- pants did not conceptually change the results (SI Appendix, Tables S6 and S16). tional reinstatement effect. Our group has previously used six CS Both experiments were approved by the Regional Ethical Review Board in presentations as a standard extinction protocol (e.g., two earlier Stockholm (https://etikprovningsmyndigheten.se/). Participants gave written experiments that investigated the effect of direct [firsthand] vs. informed consent after a short interview with the experimenter in which all observational extinction in our group) (28, 29). questions were addressed. Importantly, observational reinstatement does not contradict that other sensory modalities, especially olfactory or auditory Apparatus and Stimuli. cues of conspecifics in the animal experiment, could have con- CSs. In both experiments, two colored rectangles (562 × 762 pixels, bright tributed to the observational effect. We want to highlight that yellow and blue) were presented in the middle of the screen on a black background, serving as CSs (counterbalanced as CS+ and CS–). The CSs were other threat recovery processes, in particular renewal (recovery presented for 6 s, in a pseudorandomized order with the constraint that threat responses within novel or dangerous context), might have each CS was not presented more than two times in a row. The duration of contributed to the enhancement of threat responses by observa- the intertrial interval (ITI) between CSs was jittered between 11 and 15 s. tion of others’ US reactions. While our study was not designed USs. The US consisted of a 100-ms DC-pulse electric stimulation, which was to detect potential differences between male and female par- administered using a constant voltage stimulator (STM200; BIOPAC Systems) ticipants (when observing a US that is administered to a male applied to the participant’s right forearm through a surface electrode. A demonstrator), we want to emphasize that differences in perceiv- conductive gel (Sigma Gel) was applied between the electrode and the skin. ing responses of males and females in pain (36) and identification Before the start of each experiment, the level of the US was individually with the demonstrator model (37) might influence observational adjusted to be “unpleasant but not painful” (experiment 1: range 24 to 56 mA, mean = 36.7 mA, SD = 9.12; experiment 2: range 15.7 to 41.1 mA, reinstatement. mean = 29.39 mA, SD = 7.63), and participants were asked to rate the It is tempting to speculate that similar neural substrates of unpleasantness of the US on a scale reaching from between 1 (“not unpleas- firsthand acquisition training were active after direct and obser- ant at all”) to 10 (“extremely unpleasant”; experiment 1: range 2 to 9, vational reinstatement. This, however, remains to be tested in mean = 5.70, SD = 1.43; experiment 2: range 1 to 7, mean = 3.64, SD = 1.45). future research. We only present an exploratory investigation Observational reinstatement US. of c-fos expression in demonstrators and observers of reinstate- Experiment 1. The observational reinstatement US consisted of a video ment in SI Appendix in order to provide an initial readout of sequence depicting a male demonstrator model (age = 29 y) receiving three general, unspecific neural activity. The univariate analyses of unannounced, unpleasant electrical shocks while sitting in front of a black our preliminary dataset did not reveal any differences between screen. The demonstrator reacted to the shocks by slightly twitching the arm demonstrators and observers of reinstatement in key regions that and showing a facial expression of pain (resulting from an electric stimula- tion of the shock electrode that was visibly attached to the demonstrator’s process threats. However, finer-grained analyses might be more right wrist). The observational reinstatement USs control (no-reinstatement suitable to pick up differences in neural processes that code first- control group) in experiment 1 consisted of a video of the same male learn- hand vs. observed reaction to reinstatement USs. In support of ing model calmly sitting in front of a black screen. Acquisition, extinction, this idea, previous experiments in both rodents and humans have the observational reinstatement procedure, and the reinstatement test took revealed a partial overlap in neural activations during obser- place in a sound-attenuated cabin, illuminated by white light. vational and firsthand aversive experiences (38–45). It should, Experiment 2. The context-specific observational reinstatement US used however, be highlighted that such an overlap in neural activation the same video sequences as described for the observational reinstatement does not necessarily imply identical underlying processes (38, 42, US in experiment 1 but toned the light surrounding the demonstrator to 46, 47). Such divergent processes within overlapping central ner- a red color (red context). In experiment 2, acquisition, extinction, and the observational reinstatement procedure took place within the white con- vous pathways would further be in line with findings of spinal trol context, as described for experiment 1. The control context during the cord responses when observing others being exposed to aver- reinstatement test in experiment 2 employed this white control context as sive events that, however, are different from spinal responses to well. The context-specific testing environment consisted of another iden- firsthand aversive experiences (48). tical sound-attenuated cabin with a red light (red testing context), which In sum, our results provide experimental evidence that obser- was adjacent to the white control context. Participants were placed in each vation of reactions to aversive events interacts with learned context. association that results from firsthand aversive experiences. Procedure. Each experiment consisted of five phases: habituation, acquisi- Specifically, we found in rats and humans that observing another tion, extinction, observational reinstatement procedure, and reinstatement conspecific reacting to an aversive stimulus (US) reinstates con- test. Habituation consisted of two nonreinforced presentations of each CS. In the acquisition phase, the presentation of the CS+ coterminated with the ditioned threat responses that underlie associative memories of presentation of the US on 4 of 6 presentations in experiment 1 and on 9 of a firsthand aversive US experience. We furthermore found that 12 presentations in experiment 2. The CS–, which was never paired with the recovery of threat responses by observation of others was specific US, was presented the same number of times as the CS+ in each experiment. to the context that was observed as dangerous. Our three experi- The immediately following extinction phase consisted of six unreinforced ments provide a unique cross-species perspective, suggesting that presentations of each CS. During the observational reinstatement proce- observation of others’ distress might underlie domain general dure, participants were presented with the observational US (i.e., a video 6 of 8 | PNAS Haaker et al. https://doi.org/10.1073/pnas.2101290118 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences showing a male learning model receiving three electrical stimulations that ment was temperature and humidity controlled and maintained on a 12/12 were similar to the US that the participants experienced during acqui- light/dark cycle. Rats had ad libitum access to food and water. All conditions sition; see above). During the reinstatement procedure, the participants and procedures followed the National Research Council’s Guide for the Care themselves never received a shock. After the observational reinstatement and Use of Laboratory Animals (53), and all procedures were approved by procedure in experiment 1, a black screen was presented for 60 s. In experi- the New York University Animal Care and Use Committee. ment 2, participants waited outside the testing chamber after observational Apparatus and stimuli. All experiments were conducted using a Habitest reinstatement for 60 s before they entered the chamber and the reinstate- Linc system controlled by Graphic State 2 software. Conditioning boxes ment test was conducted. The following reinstatement test consisted of 12 (Model H10-11R-TC; Coulbourn Instruments; 30-cm width, 25-cm depth, CSs (6 CS+ and 6 CS–) presentations, which were not reinforced by the US. CS 30-cm height) were placed in sound-isolating cubicles (Model H10-24A; presentations were in randomized order and counterbalanced for the start- Coulbourn Instruments). For the habituation session and for reinstatement, ing of a CS+ or CS– presentation. In experiment 1, all 12 CSs were presented modified shuttle boxes with a transparent plexiglass dividing them were in the same context. In experiment 2, six CSs (three CS+ and three CS–) were also used (Model H10-11R-SC; Coulbourn Instruments; 50-cm width, 25-cm presented in the white illuminated context A, and six CSs (three CS+ and depth, 30-cm height). Every box had rod flooring connected to shock gen- three CS–) were presented in the red illuminated context B. erators (Model H13-15; Coulbourn Instruments), a house light, a speaker Finally, CS–US contingency awareness as well as retrospective expectancy connected to a tone generator (Model A12-33; Coulbourn Instruments), and of the US after the observational reinstatement procedure was assessed by infrared cue lights. visual analog scales (0 to 100). CS and US. The CS was a continuous tone (20-s duration; 5 kHz; 80 dB; Questionnaires. State and trait anxiety levels were assessed using the State- CStone) or the house light on for a duration of 20 s (CSlight). The US Trait Anxiety Inventory [STAI (49)] prior to the experiment. Following the consisted of an electrical foot shock (0.5 s, 1 mA). experiment, participants filled out a questionnaire assessing CS–US con- Observational reinstatement US. For reinstatement, we used unsignaled tingency awareness (e.g., if a colored square was followed by a US and foot shocks (0.5 s, 1 mA) delivered to the side of the shuttle box where which color was followed by the US). In this postexperimental question- the demonstrator was placed. During conditioning, the US coterminated naire, participants further rated retrospectively their expectancy of the US with the CS. The observational reinstatement procedures consisted of four after the observational reinstatement USs (visual analogue scale from 0 [no unsignaled US foot shocks applied to the demonstrator chamber. The expectancy] to 10 [sure about a US] for each CS and the ITI). Additionally, demonstrator rat reacted to the shocks by jumping and vocalization. PSYCHOLOGICAL AND COGNITIVE SCIENCES participants were asked to fill out a short questionnaire on their health Procedure. After being housed in pairs for at least 2 wk, on day 0, rats were status (intake of medication, etc.). habituated to the reinstatement context by placing them in the modified Skin conductance acquisition and data reduction. A pair of pregelled shuttle boxes for 30 min. In these boxes, each pair of rats was able to see Ag/AgCl electrodes was attached to the palmar side of the distal phalanges and smell each other’s odor, but they could not establish physical contact. of the index and middle fingers of the left hand. The physiological signals Twenty-four hours later (day 1), the rats were conditioned individually in were amplified and wirelessly recorded (in order to allow us to switch con- the conditioning boxes by the presentation of three pairings of CSlight-US texts in experiment 2) using a BIOPAC 150 system at a rate of 200 samples and the presentation of three pairings of CStone-US. On days 2 and 3, all per second. Data were analyzed using AcqKnowledge 4.1 software (BIOPAC rats underwent extinction training to the CSlight by presentations of 15 Systems). The raw signal was filtered (low pass: 1 Hz, high pass: 0.01 Hz), and unreinforced CSlight on each day. Extinction training took place in a mod- SCRs were manually scored for each CS trial as the base to peak amplitude ified context with a variable ITI ranging from 90 to 300 s (mean ITI was for the first response (in micro-Siemens) with base of the amplitude in the 180 s). The first CS was preceded by a 5-min acclimation period. This was in latency window from 0.5 to 4.5 s after stimulus onset using AcqKnowledge the same box as the conditioning session but with peppermint odor and a 4.1 software [as in previous experiments (50, 51)]. Trials with obvious elec- smooth black plastic floor covering the rod flooring. All sessions of all indi- trode artifacts and reactions with a rise time longer than 4.5 s (indicating vidual procedures were run for six rats at a time. On day 4, memory for artifacts from breathing) were scored as missing reactions. Amplitudes were both CSs was tested (extinction test) by presenting five unreinforced CSlight range corrected for the maximal SCR during the extinction phase to account and five unreinforced CStone in the modified conditioning boxes after 4 for interindividual variability in extinction training and logaritmized. Range min of acclimation period with a variable ITI (90 to 300 s; mean: 180 s); correction to the maximum during acquisition training did not conceptually 24 h later, each pair of rats was placed in the modified shuttle boxes, and change the results (SI Appendix, Figs. S3 and S6 and Tables S8 and S18). after 7 min of acclimation, four unsignaled foot shocks with variable ITI Data analysis. Physiological and retrospective US expectancy data were (60 to 270 s; mean: 140 s) were delivered only to the demonstrators (ran- analyzed using JASP (version 0.14). Our main analysis focused on the return domly selected) while the observer was present on the other side of the box of conditioned responses from the end of extinction toward the reinstate- separated by a transparent plexiglass. Rats were returned immediately to ment test. To this aim, in both experiments in humans, SCRs were averaged their home cages, and the animals used for immunohistochemistry were across a block of three trails in all phases of the experiment. Thereby, we euthanized 90 min after reinstatement. Post-reinstatement memory was could focus on the second half of extinction training and the first half dur- tested individually 24 h later by placing them on the modified condition- ing the reinstatement test. Importantly, our decision to include three trials ing boxes and presenting five unreinforced CSlight and five unreinforced was based on a previous study that revealed temporal stability of reinstated CStone after an acclimation period of 5 min with variable ITI (90 to 300 s; conditioned responses across three trials in humans (52). Hence, the main mean: 180 s). analysis consisted of a 2 (CS type; CS+, CS–) × 2 (phase; last block of extinc- Assessment of freezing. Videos recorded during the extinction test and tion, first block of the reinstatement test) repeated measures ANOVA. As during the post-reinstatement test sessions were analyzed for freezing by at defined previously (8), a main effect of phase would indicate a generalized least one rater who was blind to rat condition (demonstrators or observers). reinstatement, and a stimulus × phase interaction would be indicative of a Freezing was defined as cessation of all movement other than respiration. differential reinstatement effect. To this end, planned comparisons included Freezing during each CS presentation was timed with a digital stopwatch contrasts from the end of extinction to the reinstatement test in general and and presented as duration of freezing in seconds. Analogue to the human for each CS. To explore individual differences in responding during the rein- experiments, we averaged freezing behavior across the last block (two tri- statement test, correlation between individual STAI scores and differential als) of the extinction retention test and the first block (two trials) of the SCRs (CS+ − CS– trials) for the reinstatement test was calculated. The signif- post-reinstatement test. icance level was set at <0.05 for all analyses. Greenhouse–Geisser degrees Statistical analysis. The analysis of the animal data was analogous of freedom correction was used if assumptions of sphericity were violated. to the analyses of the human experiments and consisted of a Post hoc comparisons that followed upon the ANOVA were corrected for repeated measures ANOVA with two within-subject factors: CS type multiple comparisons using the Bonferroni–Holm method, correcting for (CS+extinguished, CS+nonextinguished) and reinstatement (extinction test, comparisons accordioning to our hypotheses (i.e., increases from extinction post-reinstatement) and type of learning (observer, demonstrator) as a to reinstatement within one group/context and within the reinstatement between-subject variable. Post hoc comparisons were corrected for multiple test between groups/contexts). comparisons using the Bonferroni–Holms method. Additional analyses (sug- gested by reviewing experts) that excluded outliers in the reinstatement test Animal Experiments. in each experiment did not conceptually change the results (SI Appendix, Animals. Subjects were 23 male Sprague–Dawley rats (Hilltop Laboratory Figs. S2, S5, and S7 and Tables S7, S17, and S25). Animals) weighing 225 to 400 g at the beginning of the experiments. The animals were housed in pairs during at least 2 wk before the beginning of Data Availability. Individual datasets to reproduce the figures and analyses the experiments. After conditioning, rats were single housed. The environ- and the analyses files are available at Open Science Framework (54). Haaker et al. PNAS | 7 of 8 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences https://doi.org/10.1073/pnas.2101290118 ACKNOWLEDGMENTS. This project was supported by German Research Competitiveness of Spain Grant PSI2017-84290-R. A.O. was supported by Foundation Research Grant HA 7470/1-1 (to J.H.) and German–Israeli Knut and Alice Wallenberg Foundation Grant KAW 2014.0237, European Foundation for Scientific Research and Development Young Investiga- Research Council Starting Grant 284366 (Emotional Learning in Social tor Grant 2528 (to J.H.). G.G.-B. was supported by Ministry of Educa- Interaction Project), and Swedish Research Foundation Consolidator Grant tion of Spain Grant PRX17/00284 and Ministry of Economy, Industry and 2018-00877 (Vetenskapsrådet). 1. J. LeDoux, Rethinking the emotional brain. Neuron 73, 653–676 (2012). 29. A. Golkar, J. Haaker, S. Ida, A. Olsson, Neural signals of vicarious extinction learning. 2. S. Mineka, A. Ohman, Phobias and preparedness: The selective, automatic, and Soc. Cog. Affect. Neurosci. 11, 1541–1549 (2016). encapsulated nature of fear. Biol. Psychiat. 52, 927 (2002). 30. E. Knapska, M. Mikosz, T. Werka, S. Maren, Social modulation of learning in rats. 3. A. Olsson, E. A. Phelps, Social learning of fear. Nat. Neurosci. 10, 1095–1102 (2007). Learn. Mem. 17, 35–42 (2010). 4. C. M. Heyes, Social learning IN animals: Categories and mechanisms. Biol. Rev. 69, 31. A. Nowak, T. Werka, E. Knapska, Social modulation in extinction of aversive 207–231 (1994). memories. Behav. Brain Res. 238, 200–205 (2013). 5. W. Hoppitt, K. N. Laland, Social Learning: An Introduction to Mechanisms, Methods, 32. Y. Cheng et al., Expertise modulates the perception of pain in others. Curr. Biol. 17, and Models (Princeton University Press, 2013). 1708–1713 (2007). 6. A. Olsson, E. Knapska, B. Lindström, The neural and computational systems of social 33. B. Lindström, A. Golkar, S. Jangard, P. N. Tobler, A. Olsson, Social threat learning learning. Nat. Rev. Neurosci. 21, 197–212 (2020). transfers to decision making in humans. Proc. Natl. Acad. Sci. U.S.A. 116, 4732–4737 7. M. E. Bouton, R. C. Bolles, Role of conditioned contextual stimuli in reinstatement of (2019). extinguished fear. J. Exp. Psychol. Anim. Behav. Process. 5, 368–378 (1979). 34. L. Zhang, J. Gläscher, A brain network supporting social influences in human decision- 8. J. Haaker, A. Golkar, D. Hermans, T. B. Lonsdorf, A review on human reinstatement making. Sci. Adv. 6, eabb4159 (2020). studies: An overview and methodological challenges. Learn. Mem. 21, 424–440 (2014). 35. A. Golkar, M. Bellander, A. Öhman, Temporal properties of fear extinction–does time 9. D. Hermans et al., Reinstatement of fear responses in human aversive conditioning. matter? Behav. Neurosci. 127, 59–69 (2013). Behav. Res. Ther. 43, 533–551 (2005). 36. L. Zhang, E. A. R. Losin, Y. K. Ashar, L. Koban, T. D. Wager, Gender biases in estimation 10. I. P. Pavlov, Conditioned Reflexes: An Investigation of the Physiological Activity of the of others’ pain. J. Pain 5900, 00035 (2021). Cerebral Cortex (Oxford UP, 1927). 37. A. Golkar, A. Olsson, The interplay of social group biases in social threat learning. Sci. 11. R. A. Rescorla, C. D. Heth, Reinstatement of fear to an extinguished conditioned Rep. 7, 7685 (2017). stimulus. J. Exp. Psychol. Anim. Behav. Process. 1, 88–96 (1975). 38. S. A. Allsop et al., Corticoamygdala transfer of socially derived information gates 12. M. E. Bouton, Context and behavioral processes in extinction. Learn. Mem. 11, 485– observational learning. Cell 173, 1329–1342.e18 (2018). 494 (2004). 39. S. Gallo et al., The causal role of the somatosensory cortex in prosocial behaviour. 13. M. G. Craske et al., Optimizing inhibitory learning during exposure therapy. Behav. eLife Sciences 7, e32740 (2018). Res. Ther. 46, 5–27 (2008). 40. D. Jeon et al., Observational fear learning involves affective pain system and Cav1.2 14. B. Vervliet, F. Baeyens, O. Van Den Bergh, D. Hermans, Extinction, generalization, and Ca2+ channels in ACC. Nat. Neurosci. 13, 482–488 (2010). return of fear: A critical review of renewal research in humans. Biol. Psychol. 92, 51–58 41. C. Lamm, J. Decety, T. Singer, Meta-analytic evidence for common and distinct (2013). neural networks associated with directly experienced pain and empathy for pain. 15. K. S. LaBar, E. A. Phelps, Reinstatement of conditioned fear in humans is context Neuroimage 54, 2492–2502 (2011). dependent and impaired in amnesia. Behav. Neurosci. 119, 677–686 (2005). 42. B. Lindström, J. Haaker, A. Olsson, A common neural network differentially mediates 16. R. Sjouwerman, T. B. Lonsdorf, Experimental boundary conditions of reinstatement- direct and social fear learning. Neuroimage 167, 121–129 (2018). induced return of fear in humans: Is reinstatement in humans what we think it is? 43. H. Meffert, S. J. Brislin, S. F. White, J. R. Blair, Prediction errors to emotional expres- Psychophysiology 57, e13549 (2020). sions: The roles of the amygdala in social referencing. Soc. Cognit. Affect. Neurosci. 17. J. Debiec, L. Dı́az-Mataix, D. E. A. Bush, V. Doyère, J. E. Ledoux, The amygdala 10, 537–544 (2015). encodes specific sensory features of an aversive reinforcer. Nat. Neurosci. 13, 536–537 44. A. Olsson, K. I. Nearing, E. A. Phelps, Learning fears by observing others: The neural (2010). systems of social fear transmission. Soc. Cognit. Affect. Neurosci. 2, 3–11 (2007). 18. L. Diaz-Mataix, J. Debiec, J. E. LeDoux, V. Doyere, Sensory-specific associations stored 45. T. Singer et al., Empathy for pain involves the affective but not sensory components in the lateral amygdala allow for selective alteration of fear memories. J. Neurosci. of pain. Science 303, 1157–1162 (2004). 31, 9538–9543 (2011). 46. A. Krishnan et al., Somatic and vicarious pain are represented 19. J. Haaker et al., Making translation work: Harmonizing cross-species methodology by dissociable multivariate brain patterns. eLife Sci. 5, e15166 in the behavioural neuroscience of Pavlovian fear conditioning. Neurosci. Biobehav. (2016). Rev. 107, 329–345 (2019). 47. P. L. Lockwood, M. A. J. Apps, S. W. C. Chang, Is there a ‘social’ brain? 20. T. D. Goode, S. Maren, Animal models of fear relapse. ILAR J. 55, 246–258 (2014). Implementations and algorithms. Trends Cognit. Sci. 24, 802–813 21. S. Maren, K. L. P., I. Liberzon, The contextual brain: Implications for fear conditioning, (2020). extinction and psychopathology. Nat. Rev. Neurosci. 14, 417–428 (2013). 48. A. Tinnermann, C. Büchel, J. Haaker, Observation of others’ painful heat stimulation 22. N. Sokol, P. F. Lovibond, Cross-US reinstatement of human conditioned fear: Return involves responses in the spinal cord. Sci. Adv. 7, eabe8444 (2021). of old fears or emergence of new ones? Behav. Res. Ther. 50, 313–322 (2012). 49. C. D. Spielberger, R. L. Gorsuch, R. E. Lushene, Manual for the State-Trait Anxiety 23. L. R. Halladay, M. Zelikowsky, H. T. Blair, M. S. Fanselow, Reinstatement of extin- Inventory (Consulting Psychologists Press, Palo Alto, CA, 1970). guished fear by an unextinguished conditional stimulus. Front. Behav. Neurosci. 6, 50. Y. Pan, A. Olsson, A. Golkar, Social safety learning: Shared safety abolishes the 18 (2012). recovery of learned threat. Behav. Res. Ther. 135, 103733 (2020). 24. T. D. Goode, J. J. Kim, S. Maren, Relapse of extinguished fear after exposure to a 51. J. Haaker, T. Molapour, A. Olsson, Conditioned social dominance threat: Observation dangerous context is mitigated by testing in a safe context. Learn. Mem. 22, 170–178 of others’ social dominance biases threat learning. Soc. Cognit. Affect. Neurosci. 11, (2015). 1625–1637 (2016). 25. R. W. Morris, T. M. Furlong, R. F. Westbrook, Recent exposure to a dangerous context 52. R. Scharfenort, M. Menz, T. B. Lonsdorf, Adversity-induced relapse of fear: Neural impairs extinction and reinstates lost fear reactions. J. Exp. Psychol. Anim. Behav. mechanisms and implications for relapse prevention from a study on experimentally Process. 31, 40–55 (2005). induced return-of-fear following fear conditioning and extinction. Transl. Psychiatry 26. C. Heyes, Empathy is not in our genes. Neurosci. Biobehav. Rev. 95, 499–507 (2018). 6, e858 (2016). 27. C. Heyes, J. M. Pearce, Not-so-social learning strategies. Proc. Biol. Sci. 282, 20141709 53. National Research Council, Guide for the Care and Use of Laboratory Animals (2015). (National Academies Press, Washington, DC, ed. 8, 2011). 28. A. Golkar, I. Selbing, O. Flygare, A. Öhman, A. Olsson, Other people as means to a safe 54. J. Haaker et al., Observational reinstatement. Open Science Framework. end vicarious extinction blocks the return of learned fear. Psychol. Sci. 24, 2182–2190 https://osf.io/j6sc5/view only=90e63c921e8746508f28f5268a1cbed7. Deposited 11 (2013). January 2019. 8 of 8 | PNAS Haaker et al. https://doi.org/10.1073/pnas.2101290118 Observation of others’ threat reactions recovers memories previously shaped by firsthand experiences