the viewer's perception. The Kuleshov effect, a central concept in montage theory, proposes that the emotional interpretation of neutral facial expressions is influenced by the accompanying emotional scene in a face-scene-face sequence. However, concerns persist regarding the validity of previous studies, often employing inauthentic film materials like static images, leaving the question of its existence in authentic films unanswered. This study addresses these concerns by utilizing authentic films in two experiments. In Experiment 1, multiple film clips were captured under the guidance of a professional film director and seamlessly integrated into authentic film sequences. A total of 59 participants viewed these face-scene-face film sequences and were tasked with rating the valence of neutral faces. The findings revealed that the interpretation of emotion in neutral faces is significantly influenced by the accompanying fearful or happy scene, eliciting perceptions of negative or positive emotion from the neutral face. These results affirm the existence of the Kuleshov effect within authentic film sequences. In Experiment 2, 31 participants engaged in a similar task under functional magnetic resonance imaging (fMRI). The results revealed neural correlates supporting the existence of the Kuleshov effect at a neural level. These correlates include the cuneus, precuneus, hippocampus, parahippocampal gyrus, post cingulate gyrus, orbitofrontal cortex, fusiform gyrus, and insula. These findings also underscore the contextual framing inherent in the Kuleshov effect. By seamlessly integrating film theory and cognitive neuroscience experiments, this comprehensive study provides robust evidence supporting the existence of the Kuleshov effect. It significantly enhances our understanding of film editing and its profound impact on viewers' perception.

In film theory, montage entails the combination of two or more images through film editing, creating new meanings from the foundational frames (Bordwell, 1972). From a psychological perspective, film editing significantly influences the viewer's perception of the movie, allowing them to experience emotions beyond those portrayed by the actors (Calbi et al., 2017; Carroll & Carroll, 1993; Tan, 2018). The Kuleshov effect, a renowned example of montage, aims to investigate how film editing shapes the viewer's interpretation of emotions (Barratt et al., 2016; Prince & Hensley, 1992).

The Kuleshov effect was detected through the Kuleshov experiment, which consists of a series of frames assembled in a face-scene-face sequence (Pudovkin, 1970). This sequence includes segments featuring an actor with a neutral expression, followed by an emotional scene, and concluding with a repetition of the actor displaying a neutral expression. In this experiment, viewers were tasked with assessing the emotions portrayed by the actor's face. Despite the actor's facial expression remaining consistent, the viewer frequently attributed different emotional states to these actors, and this emotional state is intricately linked to the accompanying emotional scene. For example, when the emotional scene depicts a dead girl lying in a cabin, the viewer perceives fear in the neutral face. However, when the emotional scene features an alluring woman reclining on a couch, the viewer interprets happiness in the neutral face.

Since Kuleshov's initial pilot experiment on the Kuleshov effect in the 1920s (Pudovkin, 1970), references to the Kuleshov effect in film textbooks suggest that the effect is well established (McDonald, 2022; Reisz & Millar, 1971). However, due to the vagueness and lack of control in the initial experiment, it is not clear whether the Kuleshov effect truly exists. To address this question, numerous scientific experiments have attempted to replicate the study, utilizing diverse experimental materials and paradigms throughout the past century (Barratt et al., 2016; Ekman, 1971; Prince & Hensley, 1992). Despite these efforts, the results have proven controversial regarding the existence of the Kuleshov effect (Barratt et al., 2016; Prince & Hensley, 1992).

Initially, Prince and Hensley (Prince & Hensley, 1992) attempted to replicate the Kuleshov Experiment paradigm. In Prince’s study, each sequence started with a fade-in on an actor's neutral face, followed by a cut to an object (soup, coffin, or child), and then a cut back to the actor's neutral face, concluding with a fade-out. Prince tasked the participants with discerning whether the actor's neutral face conveyed emotion. However, the majority of the participants perceived no emotion in the actor, and Prince's study did not confirm the existence of the Kuleshov effect.

In a subsequent study, Ildirar and Ewing (Ildirar & Ewing, 2018) replicated the Kuleshov effect experiment, categorizing participants into two groups: first-time viewers and experienced viewers. Ildirar edited emotional facial clips and emotional contextual clips together to investigate whether viewers constructed spatiotemporal links between the two clips, recognizing that the two clips conveyed a unified narrative. The study revealed that experienced viewers were able to construct these links, while first-time viewers could not. This finding provided confirmation of the existence of the Kuleshov effect.

While several earlier studies predominantly explored the Kuleshov effect from a film-centric perspective (Ekman, 1971; Ildirar & Ewing, 2018; Prince & Hensley, 1992), more recently investigations have employed the Kuleshov paradigm to scrutinize the contextual framing impact on facial emotion recognition from psychological perspective. Barratt et al. (Barratt et al., 2016) conducted a study that investigated the influence of emotional contexts on facial stimuli using the Kuleshov paradigm. They refined the experimental materials by incorporating neutral faces from the Karolinska Directed Emotional Faces (KDEF) picture set as facial stimuli and integrating online static or dynamic videos as varied emotional stimuli. Employing the Kuleshov sequence, participants were tasked with evaluating the valence and arousal of the actor. Barratt discovered that participants' judgments of facial expressions were indeed influenced by the emotional stimuli, thus affirming the existence of the Kuleshov effect. Subsequently, Calbi et al. (Calbi et al., 2017) enhanced the ecological validity of Barratt's experiment by exclusively utilizing dynamic images for reverse shots and introducing inter-trial intervals. Participants initially viewed a Kuleshov sequence, and then they were asked to rate the valence of the neutral face. The results revealed that the emotional context affected the emotional judgment of neutral faces, resulting in a contextdependent bias of emotion perception. In a similar vein, Mullennix et al. (Mullennix et al., 2019) rigorously manipulated the Kuleshov effect paradigm, utilizing static facial stimuli and varying presentation times to investigate the effect of visual context on the interpretation of facial expressions. After exposure to an emotional context, participants viewed a neutral face, and the results indicated that more faces were labeled as "happy" after viewing a pleasant context and more faces were labeled "sad" or "fearful" after viewing an unpleasant context. In summary, these studies collectively reveal that the assessment of neutral faces' emotional expressions is influenced by the preceding emotional context when using static faces and online emotional videos. This suggests that the Kuleshov effect can be detected with conventional psychological experiment materials.

However, when revisiting the Kuleshov effect, the significant disparity between static images and dynamic films, in accordance with the definition of film employing moving images to create continuous life experiences (Fiorelli, 2016), cannot be disregarded. Specifically, although the use of zoom-ins with static neutral faces from the KDEF picture set approximates reality (Calbi et al., 2017), it fails to capture the nuanced facial movements inherent in actual film scenes (Ambadar et al., 2005; Tan, 2018) , resulting in the omission of spatiotemporal information from dynamic faces (Krumhuber et al., 2023). Moreover, in authentic films, the continuity of backgrounds between scenes establishes a cohesive spatial context that enables viewers to perceive these scenes as unfolding within the same spatial realm (Carroll & Carroll, 1993; T. J. Smith, 2012). The replication studies predominantly rely on isolated facial expressions and emotional scenes (Calbi et al., 2017; Ildirar & Ewing, 2018; Mullennix et al., 2019), overlooking the alignment between facial backgrounds and emotional scenes present in real-world film scenarios. This departure from authentic film scenarios may potentially compromise the robustness of the validation of the Kuleshov effect. Therefore, there is a compelling need to reexamine the Kuleshov effect using high ecological film materials.

In addition to revisiting the Kuleshov effect through behavioral experiments, employing cutting-edge neuroimaging techniques can bolster its confirmation by revealing its neural correlates. This approach aligns with the neurocinematic exploration of films (Hasson et al., 2008). To our knowledge, only two studies have attempted to explore the neural correlates of the Kuleshov effect. Mobbs et al. (Mobbs et al., 2006) conducted a functional magnetic resonance imaging (fMRI) experiment that unveiled specific mechanisms involved in the Kuleshov effect, particularly regarding how emotional scenes impact neutral facial expressions. Through the juxtaposition of various emotional scenes with neutral facial expressions in a scene-face sequence, the study identified activation in distinct brain regions, including bilateral temporal pole (TP), anterior cingulate cortex (ACC), amygdala, and bilateral superior temporal sulcus (STS). This research shed light on the neural correlates of the Kuleshov effect. Similarly, Calbi et al. (Calbi et al., 2019) employed EEG and discovered that viewers rated faces as more arousing when preceded by a fearful context. The study pinpointed a high-amplitude late positive potential (LPP) in the EEG data when faces were preceded by emotional contexts, indicating a cognitive process of expectation attribution influencing how facial expressions were interpreted. This was evidenced by the activation of LPP when evaluating incongruent sequences of stimuli. However, these studies have not addressed the neural correlates of the Kuleshov effect within a face-scene-face sequence using authentic films. This gap in research makes it challenging to fully understand how the Kuleshov effect is generated during the process of film watching.

In this study, our objective is to investigate the presence of the Kuleshov effect using authentic films. In Experiment 1, we validated the existence of the Kuleshov effect with authentic films, employing film cameras to capture actors' neutral facial expressions and emotional scenes. These elements were combined into a face-scene-face sequence, ensuring consistency in the background between faces and emotional scenes in authentic film clips. Participants were then tasked with rating the emotions portrayed by the actors. Our hypothesis posited that the judgment of emotion in neutral faces would be influenced by the emotional scene (fearful, neutral, happy). In Experiment 2, we sought to explore the neural correlates of the Kuleshov effect to further validate its existence. Employing identical authentic films, we refined the Kuleshov sequence by incorporating jetters into the procedure to accommodate fMRI data acquisition. Our hypothesis postulated that the inclusion of jetters would not affect the observation of context-dependent bias. Furthermore, we aimed to observe the neural correlates of the Kuleshov effect by contrasting activation between faces. Subsequently, we examined the context-dependent bias of Kuleshov effect, investigating how brain activity elicited by faces was altered by emotional scenes.

The current study adopted the Kuleshov experimental paradigm (Calbi et al., 2017), which includes a neutral face, an emotional scene, and a neutral face presented in a face-scene-face sequence. The aim was to observe the context-dependent bias in judging neutral faces affected by accompanying emotional scenes (fearful, neutral, happy) within authentic films. In Experiment 1, we captured authentic face and scene materials and combined them in film sequences. Subsequently, these film sequences were employed in a behavioral experiment to explore the existence of the Kuleshov effect. In Experiment 2, the same films were utilized in an fMRI study to investigate the neural correlates of the Kuleshov effect.

One hundred and two healthy volunteers participated in the experiments, and all provided informed consent, which was approved by the ethics committee of the State Key Laboratory of Cognitive Neuroscience and Learning at Beijing Normal University. Participants received monetary compensation, and the experimental procedures were ethically sanctioned by the aforementioned committee.

While montage theory stands as a cornerstone in film studies, the early evidence from the original Kuleshov experiment remains ambiguous and lacks clarity (Pudovkin, 1970), leaving the existence of the Kuleshov effect an open question. Over the past three decades, despite replicated experiments suggesting the presence of the Kuleshov effect (Barratt et al., 2016; Calbi et al., 2017; Mullennix et al., 2019; Prince & Hensley, 1992), many of them relied on inauthentic film materials, such as static images. Consequently, their findings may not generalize to the existence of the Kuleshov effect in an authentic film scenario. This study aims to examine the Kuleshov effect using highly ecological films and explores whether film editing can influence viewers' perception of emotion. Additionally, we delve into an exploration of the neural correlates underpinning the Kuleshov effect to further confirm its existence from a neurocinematic perspective (Hasson et al., 2008). These evidence from the current study can further support montage theory within realistic films.

In Experiment 1, we shot authentic films, including neutral faces and emotional scenes, and integrated them into a face-scene-face Kuleshov sequence (Figure 1). Before the Kuleshov experiment, we conducted a rating experiment to establish a baseline for neutral faces. During the Kuleshov experiment, participants were asked to rate the valence of the neutral face after watching the face-scene-face sequence, revealing that the emotional scene indeed affects emotion perception from a neutral face (Figure 3A). Specifically, fearful scenes prompt viewers to perceive negative emotion from a neutral face, while happy scenes prompt viewers to perceive positive emotion from a neutral face. Meanwhile, neutral scenes did not affect viewers' perception of no expression from a neutral face. The ANOVA results showed a significant difference in valence between emotional and neutral conditions, suggesting that a viewer's interpretation of an actor's expression is significantly influenced by the emotional scene, leading to a context-dependent bias in emotional perception. Effectively controlling for the influence of prior knowledge about the Kuleshov effect ensures the reliability of our experimental results, as depicted in Figure S2. We conducted a similar Kuleshov experiment in fMRI in Experiment 2. In the behavioral results of Experiment 2, we found that the black screen inserted between emotional scenes and faces did not affect the observation of such a context-dependent bias in emotional perception (Figure 3C). Consistent with previous studies (Calbi et al., 2017; Mullennix et al., 2019), these behavioral results from Experiments 1 and 2 support the existence of the Kuleshov effect in an authentic film scenario. Notably, the Kuleshov sequence triggered higher arousal in the fearful condition compared with the neutral condition, consistent with prior research (Calbi et al., 2017). This observation indicates that the fearful condition not only initiates the context-dependent bias but also intensifies viewers' emotional experiences. Additionally, we detected activation in multiple brain regions in the contrast between Face_2 and Face_1 in each condition (Figure 4) and in the contrast between Face_2 in the fearful or happy condition and the neutral condition (Figure 5). These are the neural correlates of the Kuleshov effect, and the activation of these regions supports the existence of its effect at the neural level.

Enhancing ecological validity of the Kuleshov effect with authentic films

In contrast to previous behavioral studies, we adopted highly ecological films to ensure that participants perceived the scenes as integral components of a continuous film rather than isolated images or clips. This methodological choice, while pivotal, deviated from earlier investigations (Barratt et al., 2016; Calbi et al., 2017; Mullennix et al., 2019), which relied on facial expressions from the KDEF picture set and utilized zoom-in techniques to simulate dynamic faces from static images. However, this approach faced challenges in achieving realism, including subtle facial tremors induced by individuals remaining still, the natural occurrence of eye blinking during observations, and the static background (Ambadar et al., 2005; Ponech, 1997; Tan, 2018) , making the handling of static images as real-life film clips difficult. Moreover, earlier investigations selected face and emotional scenes separately, rather than matching them as an integrated film. As the films closely mirror reality (Lu et al., 2016; Redcay & Moraczewski, 2020), adherence to continuity-editing rules aids viewers in perceiving the film as an emulation of the real world of isolated neutral faces and emotional scenes separately, without considering the coherence of the backgrounds, led to spatial discontinuities. This approach could potentially impact the examination of the existence of the Kuleshov effect.

To address these challenges, we captured neutral facial expressions against a blue background and subsequently replaced the background with images or videos, ensuring the seamless integration of neutral faces into the film's reality (T. J. Smith et al., 2012). Furthermore, to mitigate potential influences from facial expressions, these faces were meticulously selected to maintain a neutral emotional state. In accordance with a previous study (Mullennix et al., 2019), we conducted a rating experiment, and the results for these neutral faces revealed a neutral valence (0.02 ± 0.13). Moreover, ensuring the consistency of neutral faces across emotional conditions is a prerequisite for observing Kuleshov effect bias. While some studies utilized neutral faces from the KDEF picture set, they did not assess the consistency of distinct groups (Barratt et al., 2016; Calbi et al., 2017). In the current study, the rating experiment we conducted indicated no significant differences among faces in various conditions (F1,2 = 1.23, p = 0.305). Confirming the existence of the Kuleshov effect through fMRI

Following exposure to the Kuleshov sequence, viewers attribute emotions to the neutral face (Figure 3), indicating a perceived new meaning for the second neutral face. To substantiate the existence of the Kuleshov effect, we explored the neural correlates associated with this new meaning. Contrasting activation in Face_2 and Face_1 revealed multiple activations (Figure 4, Table S1, S2, and S3), supporting the idea that the Kuleshov sequence indeed creates new meaning to the second neutral face, thus confirming the existence of the Kuleshov effect.

Specifically, our examination of the fearful condition unveiled activations in various brain regions, including the insula, cuneus, ACC, PCC, IPL, hippocampus, and OFC (Figure 4B). In contrast, both the happy condition (Figure 4C) and neutral condition (Figure 4D) displayed distinct activation patterns compared to the fearful condition, indicating the diversity in new meaning creation under different emotional contexts.

To comprehend why these regions exhibited activation upon exposure to neutral faces during the second viewing, we draw upon the notion that movie viewing encapsulates the intricacies of real-world interactions, enabling viewers to immerse themselves in the film (Redcay & Moraczewski, 2020). When scrutinizing the Kuleshov sequence, viewers engage in mental processes encompassing visual processing, theory of mind, and emotion generation. The activation of the cuneus in our findings may signifies visual processing of movie clips (Vanni et al., 2001). Simultaneously, the activation of the ACC, PCC, and IPL suggests that viewers employ theory of mind to interpret the film (Mahy et al., 2014; Mitchell & Phillips, 2015). Notably, the activation of the insula, a region associated with emotion processing, is noteworthy (Craig, 2011). The rationale behind emotion generation could be contextual processing (Zheng et al., 2022). Zheng et al. (Zheng et al., 2022) employing intracranial electroencephalogram recordings, investigated context-specific modulation among the amygdala, hippocampus, and OFC. They found that the OFC modulates the hippocampus and amygdala when perceiving emotions on a neutral face. The activation of the hippocampus and OFC in the current analysis aligns with Zheng’s study, supporting the idea that contextual processing may be the mechanism underlying the Kuleshov effect.

Unveiling the neural correlates of the Kuleshov effect bias

In our exploration of the neural correlates of the Kuleshov effect bias using fMRI, we systematically examined Face_2 activations in fearful and happy conditions compared to neutral conditions (Figure 5). Notably, the contrast between Face_2 in fearful and neutral conditions revealed activations in key brain regions, including the precuneus, PHC, PCC, precentral gyrus, MTG, AG, and FG. These activations point to the involvement of these brain regions in contextual effects and emotional responses.

As highlighted by Mobbs et al. (Mobbs et al., 2006), contextual effects play a crucial role in shaping emotional attribution, aligning with previous studies on contextual information's impact in various settings (Kumfor et al., 2018; Silveira et al., 2015). Kumfor's exploration of emotional body language as contextual information revealed a correlation between abnormal contextual influence and reduced integrity in the right PHC and left precentral gyrus. Similarly, Silveira investigated contextual information using museum labels, demonstrating activations in the precuneus. The involvement of the PCC and AG, vital components of the default mode network (DMN), reinforces their role in contextual framing during the Kuleshov effect (Ranganath & Ritchey, 2012; V. Smith et al., 2021). The activations observed in the AG and FG during fearful conditions align with findings by Quinones Sanchez et al. (Quinones Sanchez et al., 2021), emphasizing heightened involvement in facial recognition and memory retrieval in fear-inducing circumstances. Additionally, the activation of the PCC, as noted in studies on emotional word processing (Maddock et al., 2003), further supports its role in evaluating emotional meanings.

Contrasting Face_2 between happy and neutral conditions revealed activations in the right cuneus, precuneus, and CAL. These findings support the existence of contextual effects in the happy condition (Silveira et al., 2015). The correlation between gray matter volume in the right precuneus and subjective happiness, as demonstrated by Sato (Sato et al., 2015), underscores the role of the precuneus in our activation results for perceiving happiness on neutral faces.

In our pursuit of understanding the Kuleshov effect bias, we conducted a comparative analysis by contrasting the (Face_2 – Face_1) activation in the fearful condition with a similar contrast in the neutral condition, revealing distinctive activation regions (Figure 6A). Notably, the insula emerged as a key region, supporting the concept that fearful conditions heighten the perception of negative emotion from a neutral face. Simultaneously, when examining the comparison between the happy and neutral contrasts (Figure 6B), discernible activation in the precuneus, hippocampus, and FG underscores their pivotal roles in contextual processing and its effects (Silveira et al., 2015; Zheng et al., 2022). This affirmation solidifies the understanding that the observed Kuleshov effect manifests as a contextual influence.

Collectively, our results significantly contribute to understanding the neural substrates of the Kuleshov effect, emphasizing the intricate interplay between contextual effects and emotional responses.

Limitations and future directions

There are servals limitations in the current study. Firstly, it is crucial to acknowledge the potential impact of various factors on the observation of the Kuleshov effect, including viewer gender and film-watching experience (Ildirar & Ewing, 2018). While complete elimination of these influences remains challenging, we carefully balanced gender representation in our experiments and conducted a comprehensive questionnaire analysis to statistically assess participants' film-watching frequency. The results, depicted in Figure S3, indicate that watching frequency did not significantly affect the observation of the Kuleshov effect. Secondly, the role of sound in creating an authentic film experience is essential, potentially contributing to the auditory Kuleshov effect (Baranowski & Hecht, 2017). Despite our intentional removal of all sound from the current film clips, aligning with a previous study (Calbi et al., 2017), the potential interaction between visual and auditory Kuleshov effects warrants exploration in future studies. Thirdly, while the amygdala's involvement in fMRI studies of the Kuleshov effect has been documented (Mobbs et al., 2006), we did not detect its activation with FDR-correction in our study. This disparity may arise from paradigm differences, as Mobbs' study employed a scene-face sequence with static images, while ours incorporated a face-scene-face sequence with authentic film clips. Considering the constraints of creating authentic films within budgetary limits, we captured a limited number of scenes and conducted a restricted number of trials in the current study. Future research should consider expanding the number of trials to further observe the activation patterns associated with the Kuleshov effect.

In this study, our objective was to investigate the existence of the Kuleshov effect using authentic films. We produced these film clips and conducted two experiments. The results from behavioral experiment indicate that viewers perceive emotions on neutral faces after watching the Kuleshov sequence, suggesting that film editing has the capacity to alter viewer perception by introducing new meanings. Through a comparative analysis of the valence of neutral faces across different emotional conditions, we identified a context-dependent bias in the Kuleshov effect.

Specifically, viewers tend to perceive negative emotions on neutral faces after exposure to fearful film clips and positive emotions after watching happy film clips. Furthermore, our exploration extended to the neurobiological basis of the observed phenomena through fMRI experiments. The activation patterns in the insula, cuneus, precuneus, hippocampus, PCC, PHC, FG, and OFC suggest that the mechanism underlying the Kuleshov effect is contextual framing. These findings provide robust evidence supporting the existence of the Kuleshov effect at both behavioral and neural levels. Our study aligns with research utilizing the Kuleshov paradigm to investigate contextual framing (Calbi et al., 2017, 2019; Mobbs et al., 2006; Zheng et al., 2022), offering additional evidence that film editing can significantly influence viewer’s perception, evoking diverse emotional responses. Furthermore, our study contributes valuable insights to the understanding of film theory from a viewer's perception viewpoint.

The studies involving human participants were reviewed and approved by Institutional Review Board of the State Key Laboratory of Cognitive Neuroscience and Learning at Beijing Normal University. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

Zhengcao Cao: Conceptualization, Methodology, Software, Formal analysis, Investigation, Visualization, Data Curation, Writing - original draft. Yashu Wang: Conceptualization, Methodology, Investigation, Writing - review & editing. Liangyu Wu: Resources. Yapei Xie: Methodology, Investigation, Software. Zhichen Shi: Investigation, Formal analysis. Yiren Zhong:

Investigation. Yiwen Wang: Funding acquisition, Conceptualization, Supervision, Writing review & editing.

This study was financially sponsored through the Arts Project of 2019 National Social Science Fund of China (grant no.19BC041) and the Arts Project of 2023 National Social Science Fund of China (grant no.23ZD07). The authors thank them for financial support.

We express our gratitude to Professor Yong He from IDG/McGovern Institute at BNU for the assistance in the data collection process. Special appreciation is extended to the film crew at Sichuan Film and Television University for their essential support in shooting films. The authors would also thank Ran Li, Xiang Xiao, Yanlin Zhu, Dayi Zhou and Xinya Huang for their support during the study.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. of facial dynamics in interpreting subtle facial expressions. Psychological Science, 16(5), 403–410. https://doi.org/10.1111/j.0956-7976.2005.01548.x Baranowski, A. M., & Hecht, H. (2017). The Auditory Kuleshov Effect: Multisensory Integration in Movie Editing. Perception, 46(5), 624–631. https://doi.org/10.1177/0301006616682754 Barratt, D., Rédei, A. C., Innes-Ker, Å., & van de Weijer, J. (2016). Does the Kuleshov Effect Really Exist? Revisiting a Classic Film Experiment on Facial Expressions and Emotional Contexts. Perception, 45(8), 847–874. https://doi.org/10.1177/0301006616638595 Bordwell, D. (1972). The Idea of Montage in Soviet Art and Film. Cinema Journal, 11(2), 9. Calbi, M., Heimann, K., Barratt, D., Siri, F., Umiltà, M. A., & Gallese, V. (2017). How context influences our perception of emotional faces: A behavioral study on the Kuleshov effect.

Frontiers in Psychology, 8(OCT), 1684. https://doi.org/10.3389/fpsyg.2017.01684 Calbi, M., Siri, F., Heimann, K., Barratt, D., Gallese, V., Kolesnikov, A., & Umiltà, M. A. (2019).

How context influences the interpretation of facial expressions: a source localization highdensity

EEG study on the “Kuleshov