Introduction

The history of a place often tells us about the way people inhabited it and can widen the awareness of how the place and its function were involved in the greater socio-political picture of the time. In Germany, a great number of historical sites serve as memorial places or as places for informal learning, like cultural heritage sites, memorial sites, or documentation centers. This is reflected by the fact that NS memorial site visits are typical for school field trips in Germany, having the intention not only to provide pupils with a deep experience of the cruelties of the NS regime but also to foster their knowledge about this period through personal guides and visits to the accompanying exhibitions (Kuchler, 2021).

For out-of-school science labs, the model of Betz et al. (2016) describes an authentic context as an instructionally structured situation that simulates as many relevant characteristics of scientific work as possible. Both context and personal characteristics (like susceptibility) determine how authentic this simulation is perceived by its users (authentication). Perceived authenticity facilitates learning relevant psychological processes like situational interest, motivation, and information processing in turn. Similarly, for natural history and science museums, Achiam et al. (2016) assume that authentic educational programs are the product of a didactic transposition of practices of scientists from the domain of research onto the domain of education.

For authentic historical sites, the situation is different, because authenticity is typically not simulated, but given by the fact that the place is connected to relevant historical persons or events. Nevertheless, similar to models from out-of-school labs and museums, the possible impact of such a place on learning is strongly dependent on its authentication by the visitors, that is, their awareness and acceptance that they are staying at a site of historical relevance. In case of successful authentication of a historical site, several psychological processes come into play, which have been described by Zachrich et al. (2020) in their conceptual framework of encountering complex historical sources. These processes include various forms of cognitive (such as imagination, attentional focusing), affective (such as being moved or being personally attached), and physical (such as interaction, physiological response) types of engagement, ultimately cumulating in a unified experience of the historical place. The underlying assumption is that being at an authentic place that was part of the history when dealing with historical information materials facilitates learning and imagination and motivates elaborating the content (Zachrich et al., 2020).

But as far as we know, to date, there is little empirical evidence supporting these hypotheses. While quite a few studies have investigated the psychological impact of physical attributes of places (Nasar & Bokharaei, 2017; Peponis, et al., 2004; Stamps, 2011), only a small number have addressed the psychological impact of the historic dimension of places (Baron, 2012; Blaison & Hess, 2016; Lewicka, 2005, 2008).

In the case of historic places, the term “authentic” implies that there occurred an event considered to be of historic relevance, that the place was habited by a person considered historically relevant, or that it is considered prototypical of a certain historical period. Therefore, the status of authenticity is based on historical evidence, including documents, physical remains, or results of historical research. Importantly, the authenticity of a historic place is not necessarily reflected by its today’s appearance. On the one hand, much effort has been invested to keep or recreate the historical appearance and atmosphere of prominent historical places like churches, castles, or memorial sites through renovation and reconstruction. Hence, these sites are not only authentic but also look authentic, increasing the historical readability by their visitors (Baron, 2012). But, on the other hand, many historic places have undergone substantial changes, meaning that they do not provide the historic atmosphere and, in extreme cases, cannot even be recognized as authentic historic places at all.

Accordingly, to understand the psychological impact of authentic historic places, we need to disentangle the effects of their appearance in terms of physical attributes and the effects of their historic authenticity on visitors’ affective and cognitive variables. Previous research has focused on visitors’ experiences at authentic historic places utilized as informal learning settings, such as memorial sites or documentation centers. However, memorial sites, such as concentration camp memorial sites (KZ), typically hold both particular physical attributes and the historic dimension itself. That is, they are characterized both by their authenticity and their physical attributes that preserve the place’s historic atmosphere. To empirically determine the cognitive and affective impact of the authenticity dimension as such, one would need a place that holds a historic relevance that is hardly known to the public and, at the same time, offers no physical (architectural) attributes that could possibly give hints to its former history. This is the case if it is a historic place without any manmade physical attributes (like a battlefield), if the building was for the most part destroyed (like the former Berghof in Berchtesgaden), or if the building was comprehensively renovated over the years and holds a completely different function today, leaving no traces of its former history.

Therefore, the aim of the present study was to investigate the psychological impact of the historic authenticity dimension of a building whose former history is hardly known to the public and that also has no visible cues about its former history. The participants’ awareness of the building’s history was systematically manipulated by offering them different kinds of prior information. The study investigated how the awareness of being at a place with a negative NS history influenced cognitive and affective outcomes, in particular personal affect, performance in a recognition task, and evaluation of the place itself.

To begin with, in the following paragraphs, we give an overview of the state of research regarding the impact of historic sites on the perception and evaluation of the site itself, on possible affective and cognitive effects on memory performance, as well as on personal affect.

Perception and evaluation of authentic historic places

Several studies show that knowing about the historic dimension of a place influences the evaluation of the place itself and even the evaluation of the whole area surrounding the site. In a recent study, Ayton et al. (2022) found that houses in London that were marked by a Blue Plaque, indicating that it was formerly inhabited by notable men or women, achieved substantially higher selling prices than similar houses without such a Blue Plaque. This indicates that the awareness of the historical dimension of a site may strongly influence its evaluation.

However, even in the absence of visible signs, awareness of a site’s history has been demonstrated to have an effect on its evaluation. Blaison and Hess (2016) investigated the impact of knowing about different kinds of threats at a particular location on the evaluation of other places that varied systematically in their distance to that site. They found that participants were willing to pay less for a flat located near a threatening location, compared to a flat which was farther away from the threat. Furthermore, if the distance to the threat passed a critical distance, the effect inverted so that participants were willing to pay even more for the flat which was far away from the threat compared to a condition with no threat at all. Interestingly, these effects were stable regardless of whether they presented a current (landfill) or a bygone (and therefore historical) threat (a house, in which 20 years ago a murder took place). The results demonstrate that the awareness of negative historical events at a certain location has an impact on the evaluation of the whole area.

In addition, Savani et al. (2011) reported a mechanism they termed emotional residues, meaning that a large proportion of people believe that traces of previous emotions accumulate in physical spaces, thereby transferring affective states to subsequent visitors. Accordingly, in a series of studies, they found that participants tended to prefer to fill out a questionnaire in a room in which previous participants had experienced positive emotions rather than negative emotions. Furthermore, Ries and Schwan (2022) found evidence that the awareness of a place’s negative NS history may influence its evaluation even if the site is lacking the corresponding historic atmosphere; that is, participants tended to perceive the site to be more negative and with higher evoking arousal if they were aware of the place’s history compared to participants who were aware of the NS history in general or the control group. Taken together, these findings indicate that even in the absence of an authentic historic atmosphere, the awareness of a place’s history may exert an influence on the visitors’ experience and evaluation of a site, particularly with regard to the affective dimension.

Cognitive and affective effects on information acquisition

While several studies have empirically investigated the affective impact of a place’s history, to the best of our knowledge, no research to date has addressed the impact of authentic historic places on information acquisition. Yet, some conclusions about this issue can be drawn from the research regarding contextual effects on memory. In line with network models of memory, schema accounts of remembering, and educational theories of information elaboration, recent research has demonstrated that activating prior knowledge fosters memory for new information by connecting it to existing knowledge structures (van Kesteren et al., 2018, 2020). More importantly, the strength of this fostering effect depends on the congruence between prior association and the new information to be acquired. For example, in a study by van Kesteren et al. (2020), participants first had to learn pairs of pseudowords and scenes (A–B), followed by pairs of the same pseudowords and objects (A–C). In subsequent memory tests on the objects, performance was higher if scenes and objects matched thematically (e.g., bathroom—rubber duck) than if they were thematically incongruent (e.g., airplane—rubber duck). Therefore, activation of a prior knowledge in terms of a matching scene enhanced memory for the objects (although they had not been seen together). Similar results were reported in a study by Shin et al. (2021). Participants were immersed into two different VR settings (on Mars and underwater) where they had to learn several objects by finding them in the 3D scene. Afterwards, they performed a memory test on the objects either in the same or the other VR setting. Besides replicating the context-dependent memory effect of a higher retrieval if learning and testing take place in the same setting, the findings also revealed better memory for objects that semantically matched the scene in which they were learned.

Applied to historic places, these findings suggest that activating prior knowledge of NS history should facilitate encoding and retention of information associated with the NS period compared to stimuli not associated to the NS period. In addition, by becoming aware of a place’s former NS history, the perceived matching of the place with information from the NS period should be enhanced, which should also lead to a better memory of information associated with the NS period compared to information not associated with the NS period.

Furthermore, there is evidence that the valence of stimuli may also influence the subject’s memory performance (Christianson & Fällman, 1990). Christianson and Fällman (1990) found that very negative pictures were remembered better in a recognition task than neutral or positive ones, while on the other hand, they reduced memory for simultaneously presented associated words. As Ries and Schwan (2022) found that photographic material about the NS period was rated more negatively if awareness of the place’s NS history was given, one might expect that under conditions of awareness of a place’s NS history photographic material should be remembered better than under conditions of no awareness. Finally, according to Singer’s and Salovey’s (1988) mood congruity hypotheses, negative mood induced by knowing about the negative history of a place could facilitate the learning of affective stimuli that are congruent to the current personal mood. As we expected persons to be put into a negative mood when visiting dark historic places, photos with negative valence should be better memorized compared to photos with neutral valence.

Impact of historical sites on personal mood

Several studies have investigated the affective impact of visiting dark heritage places (Bilewicz & Wojcik, 2018; Brown, 2015; Nawijn & Fricke, 2015), which have an inherent historic dimension and at the same time offer a corresponding historical atmosphere through visible physical attributes. These studies show that the visit to such places, such as memorials of the victims of Nazism (Brown, 2015), the KZ memorial Neuengamme (Nawijn & Fricke, 2015), and the KZ memorial Auschwitz (Bilewicz & Wojcik, 2018), fosters the experience of negative emotions, like feelings of sadness, shock, anger, despair, and incomprehension, and may even lead to the syndrome of secondary traumatic stress.

Memorials typically show a number of physical attributes that reinstate the place’s historic atmosphere. In contrast, in many other cases, such physical attributes are lacking and possible effects on mood and cognition have to rely solely on the awareness of being at a historic place. In two experiments, Ries and Schwan (2022) investigated the impact of an everyday place holding a dark NS history on personal affect, affective judgement of related information material, and of the place itself. Two studies were conducted in a building showing no visible traces of its NS history while systematically manipulating via different audio texts the participants’ awareness of NS crimes in general and the NS crimes that had taken place in the particular building. Both studies revealed that personal mood became more negative during the studies. Furthermore, in the second study, personal mood became more negative than the control group if the participants had received information about NS history. In addition, photographic material about the NS period was rated more negatively if awareness of the building’s NS history was induced. Therefore, the findings are at least partially in line with respective findings at authentic historic sites holding both a historic dimension complemented by the corresponding atmosphere generated through physical attributes. However, the latter seems to have a stronger negative affect than a historic place offering a historic dimension without the corresponding atmosphere.

Research questions

Building upon previous research, the present study investigated the cognitive and affective effects of becoming aware of a place’s NS history at an authentic historic place lacking the corresponding historic atmosphere. We expected that the awareness of being at an authentic historic place related to the NS history should induce effects on the personal mood, on the memory performance of photographic material related or unrelated to the NS history, and on the affective evaluation of the place itself.

Analog to previous studies (Ries & Schwan, 2022), it was hypothesized that personal mood should become more negative (H1a) and personal arousal should increase (H1b) if the participants receive information about NS history in comparison to a control group that does not receive this information.

Regarding the performance in a memory task, we expected both a cognitive and an affective impact of history awareness. Regarding cognitive influence, based on research on context-dependent memory (Shin et al., 2021; van Kesteren et al., 2020), we expected that prior information about the crimes of the NS regime should activate prior knowledge about the NS topic. Additionally, becoming aware of a place’s former NS history should further enhance memory for stimuli associated with the NS period compared to stimuli not associated with the NS period. Therefore, the storing and the subsequent dependent performance in a memory task should be better for NS-associated photos compared to non-NS-associated photos in the groups with prior information about the NS crimes compared to the control group (H2).

Regarding affective influences, Christianson and Fällman (1990) found that the performance in a recognition task was better for pictures with negative valence compared to pictures with neutral valence. Therefore, we expected that the memory performance for photos with negative valence should be better than for photos with neutral valence (H3). Referring to Singer’s and Salovey’s (1988) mood congruity principle, learning should be facilitated if affective stimuli are congruent to the current personal mood. The mood of participants receiving information about the NS crimes should be more negative than the mood of the control group receiving neutral information about the buildings’ function as a research institute today. Therefore, we expected that the memory performance for photos with negative valence should be improved in conditions with prior information about the NS crimes compared to the control group (H4).

Finally, in line with previous research (Ries & Schwan, 2022; Blaison & Hess, 2016), we expected an effect of prior information on the valence and arousal ratings regarding the evaluation of the site; that is, the participants receiving information about NS crimes in general, complemented with the information about the NS crimes that happened in the particular building, should perceive the experimental room more negative and with a higher arousal compared to those participants only receiving information about NS crimes in general or the participants of the control group (H5).

Methods

The study was approved by the local institution’s ethics committee. Participants were invited via email based on the institute’s database. Participants who took part in the pre-study (described below in the materials section) did not participate in the main study. All participants gave their informed consent before taking part in the study and received eight euros for participation. At the end of each study, a debriefing took place.

Participants

A total of n = 106 participants took part in the study. Sixteen participants needed to be excluded due to prior knowledge about the history of the building. The final sample included 90 participants (control condition, C = 31, history awareness condition, H = 29, history and place awareness condition, HP = 30) between the ages of 18 and 30 (M = 23.18, SD = 2.60), all female. We decided to focus on women as participants of our study because the topic of “eugenic” sterilizations concerned predominantly women. We, therefore, expected that women will be more cognitively and affectively touched by these crimes and that women and men will react differently to respective information.

Study setting

The study took place in a former gynecological clinic, built in 1890. During the time of the NS dictatorship, at least 655 so-called “eugenic” sterilizations were conducted within the building (Bayer, 2008). The building’s function as a women’s clinic ended in 2002. Extensive renovations lasted until the year 2011 when subsequently a department of the university and a research institute moved in. Today, as the building houses several laboratory rooms, appointed with modern equipment where psychological experiments take place on a regular basis, one can hardly imagine the history of the building. Thus, the circumstances allowed us to vary the participants’ awareness about the building’s historical dimension by systematically providing them with different prior information.

Design

We used a single factorial between-subjects design, with three conditions differing in their received prior information. Group C received general information about the building’s present function as a research institute. Group H received general information about the various crimes authorized by the NS regime. Group HP received the same information as group H, complemented with information about the NS history of the building. To check the participants’ prior knowledge (manipulation check), two questions were queried at the end of the study: “Did you know beforehand that the building in which you currently are is a former women’s clinic?” and “Did you know beforehand that in the time of NS regime in the former women’s clinic forced sterilizations took place?”. Exclusion criteria were adapted to the contents of the groups’ received prior information.

Materials

The independent variable was the type of prior information offered via three audio texts. All audio texts were read by the same female speaker, consisted of 221 words, and had a total length of 2:42 min. Group C’s audio text offered general information about the structural features of the building (total number of floors and rooms) and several facts about the research institute (foundation year of the institute, total number of employees). Group H’s audio text offered information about the duration of the NS dictatorship and the crimes that took place within this time period, namely, the Holocaust, the persecution of people with different political views, the murder of people who were physically or mentally challenged, as well as the forced sterilizations carried out on women who were classified as “racially inferior.” Group HP’s audio text offered the same information as group H’s audio text and, furthermore, offered information about the NS history of the building, that is, that it used to be a women’s clinic where in the time of NS dictatorship hundreds of forced sterilizations took place. To keep the duration of both audio texts similar, the audio text of group H began by briefly mentioning parts of the instruction that all participants had received before in written form.

Further, there were 80 historical photographs used in the study because many historic sites, including memorial sites, use large sets of contemporary photos as material for information and learning (Schwan et al., 2020). The photos showed either content associated with the time of the Berlin Wall (1961–1989, photo set BW) or clinic-related content dated to the time of the NS regime (1933–1945, photo set NS). Photos of the BW set showed different states of the former border area, different states of building the BW, soldiers who guarded the border area, persons who died while attempting to escape, persons separated by the BW, border crossing stations, escape tunnels, and related content. Photos of the NS set showed persons associated with the NS health system, doctors, nurses, medical items, patients, women suspected to be forced sterilized, disabled persons suspected to be murdered, medical equipment, clinic facades, infirmaries, and related content. One of the NS photos is exemplarily shown in the following Fig. 1.

Fig. 1
figure 1

Exemplary photo showing clinic-related content dated to the time of the NS regime. Its caption was “Hospital ward of a sanatorium, approx. 1935.” Source reference: DÖW, Foto-Signatur 7908

Full size image

Forty of the pictures were target pictures, and the other 40 were distractor pictures. Twenty of the target pictures belonged to the photo NS set, and the other 20 belonged to the photo BW set. Ten of the target NS photos and 10 of the target BW photos were classified with a mean valence rather negative, the others with a mean valence rather neutral. The same was true for the distractor pictures. The photos were presented achromatic and either in a fixed horizontal (560 × 450 px) or vertical format (450 × 560 px). Each photo was accompanied by a short caption describing its content and the approximate year it was taken. The selection of the photo sets was based on a pre-study. Within the pre-study, participants (n = 20, all female) were presented a total of 160 historical photos; 80 showed content associated with BW; the other 80 showed content associated with NS. The photos were gathered from the local university’s archives as well as from research on the internet. They were presented randomly with the caption first, then the corresponding photo. The participants had to rate the valence of each of the 40 target photos along with the two possible rating choices “neutral” or “negative.” After each rating of the valence of ten photos, the participants were asked to answer a question about the pictures’ content to ensure that they kept their attention towards the pictures high. Based on the ratings, we selected 20 photos with a mean valence ranging from 0 to 0.15, with 0 indicating an absolute negative valence, and 20 photos with a mean valence ranging from 0.55 to 1 (with 1 indicating neutral valence), both for the NS photo set and the BW photo set.

A card game was used as a filler task, as it loads on the participants’ working memory in visuospatial and phonological ways. The game consisted of 15 cards showing written words and 15 cards showing pictures. By swapping its letters, each word can be transformed into a new word. The task was to build as many new words as possible and find their referring picture cards.

Measures

The study included the following computer-based tasks: rating of personal mood and arousal (at 2 points in time), valence rating of each of the 40 target pictures (this was a cover task for the learning phase, intending that the participants take a close look at each picture, while not knowing that a memory task of the pictures will take place), recognition memory of the 40 target pictures out of a set of 80 pictures, rating of the personal interest and knowledge about the topics NS and BW, and the evaluation of the site.

To rate their current personal mood, the participants were asked to indicate “In what kind of mood are you in right now?” on a 9-point Likert scale reaching from “very negative” to “very positive.” Accordingly, to rate their personal arousal, they were asked to indicate “How aroused/activated are you right now?” on a 9-point Likert scale reaching from “not aroused at all” to “very aroused.” To prevent a sexual connotation, it was pointed out in the questions on arousal that arousal is to be understood in the sense of activation.

Within the presentation phase, the participants’ cover task was to rate the valence of each of the 40 target photos. The two possible choices were “negative” or “neutral.”

In the memory task, the participants were presented with 40 target and 40 distractor pictures in randomized order. The task was to decide whether the photo had been presented to them within the study before or not. If they thought that the photo had already been presented within the study, they were instructed to press a green button with the inscription “yes” located at the right side of the keyboard; otherwise, they had to press a red button with the inscription “no” located at the left side.

To rate their personal interest in the topics NS and BW, the participants were asked to indicate “How would you rate your interest in the history of National Socialism (the Berlin Wall)?” on a 5-point Likert scale reaching from “very low” to “very high.” To rate their personal knowledge about the topics NS and BW, they were asked to indicate “How would you rate your knowledge of the history of National Socialism (the Berlin Wall)?” on a 5-point Likert scale reaching from “very low” to “very pronounced.”

To evaluate the site, the participants were asked “What kind of emotions does the room that you are currently in evoke in you?” along with a 9-point Likert scale reaching from “very negative” to “very positive” and “What amount of arousal does the room that you are currently in evoke in you?” along with a 9-point Likert scale reaching from “not arousing at all” to “very arousing.”

Procedure

The participants were randomly assigned to one of the three conditions. After being welcomed, they were guided into a room located in the basement of the building which was equipped with two chairs and two tables (see Fig. 2). The hardware was positioned on the table in the middle of the room (Table 1), namely a horizontally positioned computer screen (Iiyama ProLite, 27″, 16:9, 1920 × 1080 px), a keyboard, a mouse, and two speakers. The card game (filler task) was arranged at the other table positioned at the left wall (Table 2).

Fig. 2
figure 2

Experimental Room in the Basement of the Former Women’s Clinic

Full size image
Table 1 Means (M) and standard deviations (SD) of participants’ mood and arousal before and after receiving prior information (t1, t2) by condition (C, H, HP). Lower numbers indicate negative mood/low arousal
Full size table
Table 2 Percentage of hits and false alarms by condition (C, H, HP) and photo set (NS, BW)
Full size table

At the beginning of the study, the participants were instructed to sit down at a table where they received information about the procedure and the instructions of the study tasks. The first task was to rate their current mood and arousal (t1). After hearing the corresponding audio text, the participants had to subsequently rate their current mood and arousal again (t2). The following photo presentation phase was covered as a photo characterization task. The participants were randomly presented with the 40 target photos with the caption first, then the corresponding photo, and afterwards the task of rating its valence. After the presentation phase, the participants were instructed to sit down at another table and play a card game for 10 min (filler task). They were instructed that they would hear a gong when the time was over, indicating that they should end the card game and return to the first table. The study continued with the recognition memory task. The participants were presented with the 40 target photos and 40 distractor photos in random order and had to decide for each of the photos whether they had been presented to them within the study before or not. Again, for each photo, the participants were shown the caption first, followed by the corresponding photo. Next, they were asked to evaluate the room and subsequently had to rate their personal interest in and knowledge about the topics history of the NS and BW. Afterwards, they answered the manipulation check questions and finally, the query of some demographic variables took place. The study procedure is visualized in Fig. 3.

Fig. 3
figure 3

Procedure of the Study. C, control group, H, history awareness group, HP, history and place awareness group

Full size image

Results

To compare the personal interest about the topics NS history and BW history of all three groups, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor topic (NS history vs. BW history). The ANOVA revealed no main effect of condition, F(2,87) = 0.852, p = 0.430, ηp2 = 0.019, with post-hoc analysis revealing no significant differences in personal interest between group C and group H, p = 0.75 (MDiff = 0.26, 95% CI [− 0.29, 0.81]), group C and group HP, p = 1.00 (MDiff = 0.01, 95% CI [− 0.53, 0.55]), nor between group H and group HP, p = 0.81 (MDiff = − 0.25, 95% CI [− 0.80, 0.30]). There was a main effect of topic, F(1,87) = 23.74, p < 0.001, ηp2 = 0.214 (NS history: M = 3.90, SD = 0.97; BW history M = 3.38, SD = 1.05), with post-hoc analysis indicating that interest was significantly higher about the topic NS than BW, p < 0.001 (MDiff = 0.53, 95% CI [0.31, 0.74]). The ANOVA showed no interaction between the condition and topic, F(2,87) = 0.727, p = 0.257, ηp2 = 0.031.

To compare the self-rated knowledge about the topics NS history and BW history of all three groups, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor topic (NS history vs. BW history). The ANOVA revealed no main effect of condition, F(2,87) = 0.278, p = 0.758, ηp2 = 0.006, with post-hoc analysis revealing no significant differences in self-rated knowledge between group C and group H, p = 1.00 (MDiff = 0.08, 95% CI [− 0.43, 0.59]), group C and group HP, p = 1.00 (MDiff = 0.15, 95% CI [− 0.35, 0.66]), nor between group H and group HP, p = 1.00 (MDiff = 0.08, 95% CI [− 0.44, 0.59]). Again, there was a main effect of the topic, F(1,87) = 80.826, p < 0.001, ηp2 = 0.482 (NS history: M = 3.18, SD = 0.84; BW history M = 2.44, SD = 0.94), with post-hoc analysis indicating that knowledge was significantly higher about the topic NS than BW, p < 0.001 (MDiff = 0.74, 95% CI [0.57, 0.90]). The ANOVA showed no interaction between condition and topic, F(2,87) = 2.431, p = 0.094, ηp2 = 0.053.

Personal mood (H1a)

To compare the personal mood of all three groups, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor time (t1 vs. t2). The ANOVA revealed a main effect of condition, F(2,87) = 10.530, p < 0.001, ηp2 = 0.195, with post-hoc analysis revealing a significant difference in mood between group C and group H, p < 0.001 (MDiff = 1.05, 95% CI [0.37, 1.73]), as well as group C and group HP, p < 0.001 (MDiff = 1.14, 95% CI [0.47, 1.82]), but not between group H and group HP, p = 1.00 (MDiff = 0.09, 95% CI [− 0.60, 0.78]). Further, there was a main effect of time, F(1,87) = 149.476, p < 0.001, ηp2 = 0.632, with post-hoc analysis indicating that mood was significantly more negative at t2 than at t1, p < 0.001 (MDiff = − 1.90, 95% CI [− 2.21, − 1.59]). The ANOVA showed an interaction between condition and time, F(2,87) = 22.749, p < 0.001, ηp.2 = 0.343, indicating that while there was no difference between the mood of different groups at t1, the mood of groups H and HP was more negative compared to group C at t2, which is in line with H1a. Descriptive data of mood and arousal is shown in Table 1

Personal arousal (H1b)

To compare the arousal of all three groups, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor time (t1 vs. t2). The ANOVA revealed neither a main effect of condition, F(2,87) = 0.083, p = 0.921, ηp2 = 0.002, with post-hoc analysis revealing no significant difference in arousal between group C and group H, p = 1.00 (MDiff = − 0.09, 95% CI [− 0.97, 0.79]), group C and group HP, p = 1.00 (MDiff = 0.06, 95% CI [− 0.82, 0.93]), nor between group H and group HP, p = 1.00 (MDiff = 0.15, 95% CI [− 0.74, 1.03]), nor a main effect of time, F(1,87) = 0.371, p = 0.544, η2 = 0.004, with post-hoc analysis revealing no significant difference in arousal between time t1 and t2, p = 0.54 (MDiff = 0.09, 95% CI [− 0.20, 0.38]). The ANOVA revealed an interaction between condition and time, F(2.87) = 7.675, p = 0.001, ηp2 = 0.150, which is in line with H1b. Descriptively, arousal of group C decreased while the arousal of groups H and HP increased from t1 to t2, but none of the pairwise comparisons was significant.

Memory of the photos (H2, H3, and H4)

First, we computed the number of hits and false alarms of every participant (see Table 2).

We then calculated participants’ sensitivity measure d′ as indicator for the participants’ recognition performance. As d′ did not pass normal contribution, we transformed it in two steps. First, we determined its maximum to be 1 and its minimum to be − 1; afterwards, we arcsin transformed d′.

To compare the participants’ sensitivity in the recognition task, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor photo set (NS vs. BW). The ANOVA revealed no main effect of condition, F(2,87) = 0.132, p = 0.877, ηp2 = 0.003, with post-hoc analysis indicating there were no significant differences in the sensitivity between the groups C(M = 0.58, SD = 0.42) and H(M = 0.59, SD = 0.37), p = 1.00 (MDiff = − 0.01, 95% CI [− 0.26, 0.24]), the groups C and HP(M = 0.55, SD = 0.55), p = 1.00 (MDiff = 0.04, 95% C [− 0.21, 0.28]), nor between the groups H and HP, p = 1.00 (MDiff = 0.05, 95% CI [− 0.20, 0.30]). The ANOVA did reveal a main effect of the photo set, F(1,87) = 21.405, p < 0.001, ηp2 = 0.197, indicating that sensitivity for NS photos (M = 0.88, SD = 0.48) was significantly higher compared to BW photos (M = 0.61, SD = 0.48) over all groups, p < 0.001 (MDiff = − 0.28, 95% CI [− 0.40, − 0.16]). In contrast to H2, the ANOVA revealed no interaction between the condition and photo set, F(2,87) = 0.113, p = 0.893, ηp2 = 0.003.

To analyze whether the valence of the photos affected the performance in the recognition task, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor valence (neutral vs. negative). As the dependent variable, we compared d′ for the photos categorized with negative valence to d′ of the photos categorized with neutral valence due to the pre-study. In contrast to H3, the ANOVA revealed neither a main effect of condition, F(2,87) = 0.116, p = 0.890, ηp2 = 0.003, with post-hoc analysis indicating there were no significant differences in the sensitivity between the groups C and H, p = 1.00 (MDiff = − 0.05, 95% CI [− 0.33, 0.23]), the groups C and HP, p = 1.00 (MDiff = − 0.04, 95% CI [− 0.32, 0.24]), nor between the groups H and HP, p = 1.00 (MDiff = 0.01, 95% CI [− 0.27, 0.29]), nor a main effect of photo valence, F(1,87) = 0.167, p = 0.684, ηp2 = 0.002, with post-hoc analysis indicating no significant differences in the sensitivity between photos with neutral and negative valence, p = 0.684 (MDiff = 0.03, 95% CI [− 0.10, 0.16]). Furthermore, in contrast to H4, the ANOVA revealed no interaction between condition and valence, F(2,87) = 1.649, p = 0.198, ηp2 = 0.037. Descriptive data is shown in the following Table 3.

Table 3 Means (M) and standard deviations (SD) of participants’ d′ by valence due to the pre-study classification (neutral, negative) and condition (C, H, HP)
Full size table

To make sure that the current treatment did not influence the rating of the photo’s valence, we conducted a 3 × 2 Mixed Model ANOVA with the between-subjects factor condition (C vs. H vs. HP) and the within-subjects factor photo set (NS vs. BW). As the dependent variable, we used the mean rating of the photo’s valence collected within the main study’s presentation phase, noting the restriction that ratings were only available for the target pictures. The ANOVA revealed no main effect of condition, F(2,87) = 0.117, p = 0.889, ηp2 = 0.003, with post-hoc analysis revealing no significant difference in the photo ratings between group C and group H, p = 1.00 (MDiff = − 0.02, 95% CI [− 0.10, 0.07]), group C and group HP, p = 1.00 (MDiff = 0.00, 95% CI [− 0.09, 0.09]), nor between group H and group HP, p = 1.00 (MDiff = 0.02, 95% CI [− 0.07, 0.10]). However, the ANOVA showed a main effect of the photo set, F(1,87) = 14.253, p < 0.001, η2 = 0.141, with pairwise comparisons revealing that the pictures of the NS photo set were rated significantly more negative than the pictures of the BW photo set, p < 0.001 (MDiff = − 0.07, 95% CI [− 0.10, − 0.03]). The ANOVA revealed no interaction between the condition and photo set, F(2.87) = 0.149, p = 0.862, ηp2 = 0.003. Descriptive data is shown in the following Table 4.

Table 4 Means (M) and standard deviations (SD) of participants’ ratings of the target photos valence due to the presentation phase of the main study by photo set (NS, BW) and condition (C, H, HP). Lower numbers indicate negative valence
Full size table

Evaluation of the site (H5)

To test our hypotheses of perceived valence and arousal evoked by the site, we conducted ANOVAs with the factor condition (C vs. H vs. HP). Descriptive data of participants’ ratings of the site’s valence and evoking arousal are shown in Table 5.

Table 5 Means (M) and standard deviations (SD) of participants’ ratings of the site by condition (C, H, HP)
Full size table

Regarding the valence ratings of the site, the ANOVA indicated significant differences due to condition, F (2,87) = 5.489, p = 0.006, ηp2 = 0.112, which is in line with H5. Post-hoc analysis revealed no significant difference in the valence ratings of the site between group C and group H, p = 0.80 (MDiff = 0.35, 95% CI [− 0.42, 1.12]), nor between group H and group HP, p = 0.12 (MDiff = 0.67, 95% CI [− 0.11, 1.44]), but a significant difference between the groups C and HP, p < 0.01 (MDiff = 1.02, 95% CI [0.26, 1.78]), indicating that group HP rated the site more negatively than group C. Regarding the arousal ratings evoked by the site, the ANOVA indicated significant differences due to condition, F (2,87) = 3.660, p = 0.03, ηp2 = 0.078, which is again in line with H5. Descriptively, group HP rated the site with higher arousal compared to both groups H and C, but none of the pairwise comparisons was significant. With post-hoc analysis revealing no significant difference in the arousal ratings of the site between group C and group H, p = 1.00 (MDiff = − 0.02, 95% CI [− 1.21, 1.16]), between group C and group HP, p = 0.06 (MDiff = − 1.15, 95% CI [− 2.32, 0.03]), nor between the groups H and HP, p = 0.07 (MDiff = − 1.12, 95% CI [− 2.31, 0.07]).

Discussion

The aim of the present study was to investigate the impact of becoming aware of a place’s historic dimension both on a cognitive and an affective level. Previous research has focused on visitors’ experiences at authentic historic places utilized as informal learning settings, such as memorial sites or documentation centers. But as these historic sites are usually characterized both by their authenticity and their physical attributes that preserve the place’s historic atmosphere, it is an open question if it is one of these attributes or the interplay of both fostering the effects, cognitive and affective, on visitors.

The present study took up this issue by disentangling the effect of history awareness from the effect of a site’s current atmosphere generated by physical attributes from the building’s history. In addition, two forms of historical awareness were investigated, namely, receiving information about NS crimes in general or additionally being informed about NS crimes that happened in the building in which the study took place. In particular, the impact of historical awareness on the memory performance of historic photos was investigated. The photos showed either content associated with clinics in the NS period or content associated with the period of the building of the Berlin Wall and were either negative or neutral in content. We expected that the memory performance of the groups that received information about the NS crimes should be better for photos of the NS photo set compared to photos of the BW photo set because the learning of photos congruent to the already activated cognitive nodes (nodes of clinics and the NS period) should be facilitated. The facilitated learning could in the next step foster a better memory performance. Contrary to hypothesis H2, the results showed no differences in memory performance due to prior information. Instead, we found that memory performance in all groups was better for photos of the NS photo set compared to the BW photo set. A possible explanation for this finding may be that the rated interest and knowledge were higher for NS history compared to the BW history. Therefore, attention towards the NS photos during the photo presentation phase may have been high in all groups, regardless of the prior information that was given at the beginning of the study. In turn, this may have facilitated the encoding and retention for photos of the NS photo set.

It was also hypothesized that memory for photos with negative valence should be better than for photos with neutral valence, and furthermore, the memory performance for photos with negative valence should be improved in the groups that received information about the NS crimes, compared to the control group. Contrary to hypotheses H3 and H4, results showed neither an effect of photo valence nor an effect of prior information on the performance in the memory task. Exploratively, we analyzed the participants’ ratings of the photos’ valences collected in the main study’s photo presentation phase and found that irrespective of the different prior information, the participants rated the target photos of the NS photo set more negatively than the target photos of the BW photo set.

These results are neither in line with previous findings reporting that performance in a memory task was better for pictures with negative valence compared to pictures with neutral valence (Christianson & Fällman, 1990) nor with Singer’s and Salovey’s (1988) mood congruity principle, which assumes that learning should be facilitated if affective stimuli are congruent to the current personal mood. An explanation for this discrepancy might be that the recognition task was too easy. Percentual hit rates of the different groups reached from 89.85 up to 95.65%, indicating a ceiling effect.

In addition, results showed that both types of historical awareness had a negative effect on personal mood. That is, the participants who received information about the NS crimes became significantly more negative than the control group. Regarding personal arousal, the results showed a significant interaction between time and condition, but pairwise comparisons showed no significant differences. Descriptively, the arousal of the participants who received information about the NS crimes increased, while the arousal of the participants in the control group decreased. These results are both in line with hypotheses H1a and H1b and with previous findings reporting that participants’ moods became more negative, and arousal descriptively increased after receiving information about the NS history compared to a control group (Ries & Schwan, 2022). Also similar to the present study, Ries and Schwan (2022) found no evidence that the decrease in mood and increase in arousal was particularly pronounced when the participants became aware of being at a place where NS crimes happened.

Finally, we investigated the impact of becoming aware of being at a historic site on the evaluation of the site itself. The results showed the site itself was perceived less positively and descriptively evoked more arousal when participants were aware of its NS history. These findings are both in line with hypothesis H5 and with previous findings reporting partial evidence that participants who knew about the NS history of a building rated the site more negatively and with a higher evoked arousal compared to a control group (Ries & Schwan, 2022).

To sum up, the study revealed no evidence for cognitive or affective effects induced by history awareness within a memory task; that is, there were no differences found due to the different prior information regarding the performance in a memory task with congruent versus incongruent historic photos.

Furthermore, the study revealed a negative effect of history awareness on personal mood. Consistent with previous findings (Ries & Schwan, 2022), it made no difference whether the participants were only aware of the NS history in general or if they additionally were aware of the particular NS history of the building. Previous studies at KZ memorial sites have found intense negative feelings of visitors (Bilewicz & Wojcik, 2018; Brown, 2015; Nawijn & Fricke, 2015). The main difference lies in the visible characteristics of the places: While the present studies were conducted at a seemingly neutral place giving no hints for the crimes that happened within it, KZ memorial sites make NS crimes tangible through the site’s atmosphere, their architectural features, and authentic material artifacts. We therefore speculate that becoming aware of the history of a place is not sufficient enough in itself but has to be embedded in a context that reflects to a certain degree the authentic atmosphere of the historic place. However, the study also revealed that becoming aware of a place’s history fosters that the room itself was perceived to be significantly more negative and evoked more arousal.

Nevertheless, there are certainly some empirical limitations to the study. In the present study, affective reactions were determined via the two dimensions of mood and arousal, while research at KZ memorial sites has shown that visitors experience a broad range of different emotions (Brown, 2015). Therefore, future studies should try to investigate the spectrum of possible affective reactions in more detail. Also, the present studies took place in a room with no indication of its history and presumably no atmospheric resemblance to its former history. To gain more insights into the interplay of a place’s atmosphere and its historic dimension, in future studies both aspects should be systematically crossed. Regarding memory and learning, we cannot be sure if the lack of cognitive and affective effects did not perhaps result from the circumstance that the memory task had been too facile. Accordingly, future studies should use more complex learning materials, such as texts, and should also choose more sensitive measures of memory, such as free recall instead of recognition. Also, one could argue that d′. is not the most adequate operator for memory performance regarding the impact of history awareness. Therefore, future studies might operate with additional quantitative and qualitative measurements that address memory and learning, including knowledge tests and indicators of cognitive load. In addition, to determine the generalizability of the findings, future studies should investigate various kinds of authentic places and might also analyze the impact of participants’ age, gender (the present study focused exclusively on women), and place attachment (e.g., local inhabitants vs. visitors).

Taking these caveats into account, to the best of our knowledge, this is one of the first studies that has addressed the impact of historic awareness on memory, knowledge acquisition, and mood. Moreover, the topic is not only of relevance and interest within psychology but also for other disciplines and practical implications within the context of museums and memorial sites, such as history education, remembrance, reconstructions, or architecture. In the light of the present findings, we speculate that becoming aware of the authentic historical dimension of a place may be a necessary but not a sufficient condition for exerting an effect on cognitive processing and information acquisition. Instead, it seems to be that a context that reflects to a certain degree the authentic atmosphere of the historic place is an important requirement for authenticity to influence learning. As an example, KZ memorials as well as popular authentic cultural heritage sites typically show clearly visible material traces of its history and at least partly reproduce the original atmosphere. Therefore, in terms of further research, the interplay of historical awareness and atmosphere at authentic sides should be addressed in future empirical studies.

In terms of practical education implications, the lack of historical atmosphere and material remains might be compensated by the provision of applications displayed on mobile devices, like photographs of former historical conditions or Augmented Reality representations that allow for visually and/or auditively overlaying the present state of a site with digital reconstructions of its historic conditions (Amakawa & Westin, 2017; Harley et al., 2016; Price et al., 2016). Finally, another way to promote students’ or other visitors’ learning about (local) history is to combine different learning opportunities, such as lectures about the past, visiting sites, and performing one’s own historical research at authentic places, as emphasized by “dig where you stand” approaches or citizen science initiatives (Stefaniak et al., 2017).