For example, they proposed that people with unusual patterns of non-verbal behavior or physical characteristics outside the social norm e. Thus, McAndrew and Koehnke argued that this unpredictability leads to uneasy feelings about these non-conformist people and to ambiguity about how to behave and how to judge them. Another area of research offering particularly useful ideas to understanding creepiness is human-computer interaction. Within this field, scholars and practitioners Seyama and Nagayama, ; Walters et al. The drop of acceptance i. As a result, people feel uneasy about interacting with such robots and virtual agents, but they also feel ambiguity about how to behave and how to judge them.
Above and beyond creepiness in interpersonal situations and in human—computer interaction, Tene and Polonetsky provided an excellent overview of creepiness elicited by novel technologies and technologies used in novel situations. For example, they describe personalized analytics i. An example of creepy personalized analytics are algorithms predicting whether there is a pregnant person in a household and when the person will give birth see Tene and Polonetsky, , thus personalized advertising for baby products is provided.
This might be useful for organizations selling baby products, but people who are confronted with such personalized advertisements might feel uncomfortable because they do not really know why websites they visit are suddenly providing them with suggestions on where to buy baby products. Another example of a creepy situation is a situation where social listening is applied. Tene and Polonetsky describe a situation where a person having problems with their TV calls their friend for help. However, the user has no idea how the company knew there was an issue with the TV. It could be that the TV producing company monitors all problems with their TVs and calls users having severe issues.
It could also be that the user assumes that the company has monitored their call with their friend. Considering all the aforementioned work on creepiness, creepiness seems to be elicited by unpredictable people, situations, or technologies, and it seems that this induces rather unclear feelings of discomfort paired with uncertainty about how to behave during a creepy situation or with a creepy person or technology. Therefore, we can define creepiness as a potentially negative and uncomfortable emotional response paired with perceptions of ambiguity toward a person, technology or even during a situation.
Furthermore, we can preliminarily assume that creepiness consists of two subdimensions, emotional creepiness and creepy ambiguity, and preliminarily define emotional creepiness as a rather unpleasant affective impression elicited by unpredictable people, situations, or technologies and creepy ambiguity as a lack of clarity on how to act and how to judge in such a situation.
Previous studies capturing creepiness have not used a consistent creepiness scale, nor did they investigate the psychometrical properties of their creepiness measures. In all of these studies, creepiness was measured with a single item e. This might be useful to capture the general creepiness of a situation, but it makes it hard to determine reliability. Moreover, this kind of measure does not distinguish emotional parts of creepiness from the ambiguity parts. Therefore, it is harder to discern what exactly about the situation has led to high creepiness values.
Distinguishing between emotional creepiness and creepy ambiguity might help to understand which part of a situation needs to be adjusted to decrease creepiness. At this point it is necessary to differentiate between creepiness and eeriness Ho and MacDorman, ; Burleigh et al. Although both creepiness and eeriness describe similar negative reactions to strangeness and unfamiliarity and both are associated with emotions such as disgust, shock, and nervousness Ho et al.
Therefore, eeriness is nearly exclusively applied to research comparing different humanoid robots or versions of virtual characters, which is further supported considering that the eeriness scale developed by Ho and MacDorman seems to be tailored to such research e. In contrast, the concept of creepiness can be applied to a broader range of situations: It relates to situations where people find themselves in interpersonal situations McAndrew and Koehnke, or to situations where people interact with novel technologies Tene and Polonetsky, Second, although creepiness should consist of an emotional response similar to the emotional response within sensations of eeriness, there should also be a cognitive response to the ambiguity and unpredictability of the situation, which is not a part of the eeriness concept Ho et al.
Third, eeriness seems to be especially related to fear Ho et al. As a result, creepiness with its ambiguous nature, where people potentially cannot point the finger to what exactly bothers them in a certain situation e. Fourth, there is some initial evidence by Ho et al. Consequently, we argue that creepiness and eeriness are distinct in a way that creepiness may be applied to a broader range of situations and that creepiness should not only consists of an emotional but also of a more cognitive response to creepy situations.
In the following sections, we will describe the scale development approach for the CRoSS in which we closely followed recommendations by Hinkin The scale development process consisted of four studies. In the first study, we collected data from an American sample on our initial set of items to carry out an exploratory factor analysis EFA to enhance understanding of the dimensions of creepiness. Please note that Fabrigar et al. Additionally, we reduced the amount of items to increase efficiency of the scale. In the second study, we collected data from a German sample to apply a confirmatory factor analysis CFA to support the factors found in the first study.
In the third study, we examined the convergent using privacy concerns, transparency, controllability and computer anxiety and divergent validity using extraversion and conscientiousness of the CRoSS. In the last study, we used the CRoSS in a field experiment to provide further validity evidence, to show that it is sensitive to experimental manipulations based on theoretical assumptions, and to show that the CRoSS is useful in situations extending beyond the use of technology. In this study, experimenters male vs. The authors consulted the literature for studies on creepiness to obtain an overview of existing theories and measurement models of creepiness.
Based on the research of Mori , Mori et al. These items were intended to capture the facets of creepiness as inferred by prior research i. The six items that should capture emotional creepiness were written to represent unclear and queasy feelings toward a situation, whereas the eight items for creepy ambiguity were written to reflect uncertainty on how to judge a situation and how to behave during a situation. Items were generated in German, translated to English, sent to a native English-speaking proofreader, translated to German and checked for coherence with the original items.
Concerning the response format, we consulted research by Lozano et al. According to their findings, a seven-point rating scale should provide a good foundation for obtaining adequate psychometric properties of a newly developed scale. Therefore, we decided to use a seven-point rating scale from 1 Strongly Disagree to 7 Strongly Agree.
- Creepiness Study - Results.
- Creepiness–Convenience Considerations Change Over Time.
- Save or Share this Report;
Initial items in German and English, proposed dimensions of these items, and results of the exploratory factor analysis. The MTurk participants received a small amount of money for participating.
On the Science of Creepiness
For this study as well as for all of the following studies, participants were informed that they provide consent and agree that their data will be used for research purposes by continuing the respective study. During Study 1, participants watched a video where a situation similar to one of the creepy situations described by Tene and Polonetsky was shown.
In this video, a person sits in front of a computer screen using a word processing software when suddenly the computer produces an audible error signal; the person uses the mouse but nothing happens i. As a result, the person turns off the computer. Following, the person tries to restart the computer, but it does not turn on again. Afterward, the person reaches for their smartphone and starts texting a friend for help.
In the video, the screen of the smartphone is visible so participants can read what the person is writing. It is also made clear that the person is writing to a friend, because there is already a texting history clarifying that they know each other i. Once it is clear that the person asks a friend for help but before the message is sent to the friend, the person receives a call from an unknown number and starts acting confused over the call e.
The person presses the button to answer the call. This is Chris from Computer Solutions. We heard that you are having problems with your computer? You were writing something but suddenly you could not move the mouse anymore and now the computer is not turning on again? Fortunately, this is a common problem with your computer series, I can help you fix this right now.
We dubbed this phone call to ensure that participants can hear it loud and clearly. Then, the video fades out without any further information. After watching the video, participants completed the 14 initial CRoSS items and provided demographic information. Additionally, participants had to describe what happened during the situation as a manipulation check. The manipulation check was to ensure whether participants had watched the video attentively and to explore if they perceived the situation displayed in the video as ambiguously as intended e.
As a first step of Study 1, we analyzed the open-ended manipulation check question that asked participants for a description of what has happened during the situation. These explanations showed that participants watched the video and that the video generated a variety of ideas about what has happened during the situation. These commentaries showed that the video evoked reactions varying from a neutral description of the situation to the fear of privacy invasion and a hacker attack. Based on the theoretical assumptions Tene and Polonetsky, ; McAndrew and Koehnke, , the two proposed scales of creepiness should be non-orthogonal i.
We chose oblique rotation in order to allow the two proposed scales of creepiness to be correlated Fabrigar et al. To assess dimensionality, we used three criteria: The Kaiser-Guttman criterion i. Results indicate a two-factorial solution accounting for 61 percent of variance. We then analyzed the items regarding potential item removal Hinkin, For the remaining ten items, we conducted another principal component analysis with oblique rotation that resulted in two factors explaining 68 percent of variance.
In line with our initial idea about the potential dimensionality of creepiness, results showed a two-factor solution with five items on each factor. The first factor reflected emotional creepiness, with the items capturing an emotional response to a potentially creepy situation. The second factor reflected creepy ambiguity, with items describing insecurity about how to behave during the situation and how to judge the situation. Furthermore, we conducted a scale reliability analysis to ensure reliability of the entire scale and the two subscales. For the entire scale we found a good reliability cf.
For the next step of the scale development, the goodness of fit of the resulting factor structure needs to be assessed Hinkin, As such, we followed suggestions by Hinkin regarding the required sample size for a CFA and collected data from German participants in an online study. Participants were recruited through social media, in psychology and economics courses at a German university, and on an online survey platform on which researchers take part in online surveys in exchange for other people to take part in their surveys.
Three participants were excluded because of technical problems, and one participant was excluded because he stated that he did not take the study seriously. During the study, participants watched the same video as in the first study and afterward responded to the ten CRoSS items and to demographic questions. Similar to Study 1, participants had to describe what happened during the situation as a manipulation check. The only difference between the two samples was that no participant in the German sample questioned the abilities of the user in the video.
This shows support for the fact that the situation in the video was also perceived ambiguously by the German participants. Since this sample was collected in Germany, where one might expect different results for the factors and reliability of the CRoSS compared to the American sample from Study 1, an EFA with oblique rotation was conducted for the items. These results tentatively indicate that there are no substantial differences in the results of the EFA, the scale reliabilities, and the reactions to the video regarding creepiness of the American sample from Study 1 and of the German sample from Study 2.
Resulting model of the confirmatory factor analysis in Study 2. Numbers represent standardized loadings. Creepiness consisted of the two factors emotional creepiness and creepy ambiguity, both loading on a common underlying factor called Creepiness. To conclude, the CFA indicated that the two factors emotional creepiness and creepy ambiguity which both loaded on the same general creepiness factor represented the data well. Accordingly, the next step of scale development is to gather evidence of construct validity. For this purpose, the next two sections cover an online and a field experiment to examine validity of the CRoSS.
For the development of a new scale it is important to show that it is measuring a meaningful construct Hinkin, Therefore, it is necessary to demonstrate correlations with other relevant constructs convergent validity and, at the same time, distinguishability from unrelated constructs divergent validity. This scale development step is especially important in the case of creepiness, which research has just started to examine cf.
In an attempt to support convergent validity of the CRoSS, we propose correlations between creepiness and the constructs privacy concerns, computer anxiety, transparency, and controllability. Below, we provide theoretical support for each of the proposed correlations. Privacy concerns are an important variable to measure feelings of privacy invasion through novel technologies Smith et al.
When people hold privacy concerns, they are under the impression that their personal data might be collected without their knowledge, that they have no control about which data are collected, that there might be errors in the data collection, and that personal data might be misused Shin, ; Smith et al. Consequently, privacy concerns can lead to less trust in the organizations which elicited these concerns Smith et al.
This can detrimentally affect important organizational outcomes such as, applicant reactions, provision of personal information, and online sales revenue Phelps et al. Shklovski et al. For instance, they describe the invasion of privacy through smartphone apps. If people perceive privacy concerns because an app requests access to their pictures and contacts, although the app is for a game that has nothing to do with pictures or contacts, they can get a feeling that this somehow feels wrong Shklovski et al.
As such, we propose that creepiness relates to privacy concerns as both feelings can be elicited through uncontrollable situations see also Phelps et al. In fact, privacy concerns seem to decrease if people have at least the impression that they are more in control of their data Phelps et al. More precisely, if people are concerned about their privacy, it can induce uneasy feelings Powell, Therefore, we propose:.
Computer anxiety can be defined as an uncomfortable feeling when interacting with a computer or when there is the possibility that one has to use a computer Chua et al. Accordingly, creepiness relates to computer anxiety as people who are generally more anxious when it comes to interacting with a computer might also be people who will experience higher levels of creepiness when it comes to technology-related situations.
Thus, we propose:. Transparency of a situation is given if people understand what is going on during this situation Truxillo et al. In contrast, if people conceive that there is something shady about the situation or that they do not see through a situation, this reduces transparency. It is likely that situations that are not transparent are also creepy because if a situation is not instantly clear, people might come up with several possibly wrong explanations about this situation, thus increasing ambiguity see also Studies 1 and 2.
Conversely, if the providers of the advertisement made clear from where they received their information, this situation would be less ambiguous, more predictable, and thus less creepy. We therefore propose:. Hypothesis 1c: Creepiness is negatively correlated to transparency. This relation might be more pronounced for creepy ambiguity. The more people perceive that they are able to influence a situation, the more they think it is controllable Ajzen, For example, people trying to avoid personalized advertising might be successful so long as their friends and family do not spend time on the internet.
When a friend allows apps to access contact information on their smartphones, advertisement can become personalized for the person who originally tried to avoid it Shklovski et al. Consequentially, these people no longer feel in control of personalized advertising because no matter what they do, advertisers will be able to obtain information about them that they will use to personalize advertisements. This lack of control might also lead to unpredictability, as it is less possible to influence the future within uncontrollable situations.
As such, perceived control also relates to creepiness as decreased predictability increases the creepiness of situations Tene and Polonetsky, ; McAndrew and Koehnke, Furthermore, low controllability might especially be related to emotional aspects of creepiness, as low controllability seems to relate to negative affective impressions cf.
We thus propose:. Hypothesis 1d: Creepiness is negatively correlated to controllability. This relation might be more pronounced for emotional creepiness. To provide evidence for divergent validity, we chose the personality dimensions extraversion and conscientiousness, as both are expected to be unrelated to creepiness. In the case of extraversion, it should not matter if a person is especially outgoing or rather reserved in judging the creepiness of a situation. In the case of conscientiousness, a person who is rather lazy should be equally influenced by a creepy situation like a person who closely keeps track of their daily schedule.
Hypothesis 1e: Creepiness is not or at least to a lower extent in comparison to the convergent validities correlated to extraversion. Hypothesis 1f: Creepiness is not or at least to a lower extent in comparison to the convergent validities correlated to conscientiousness. Three participants were excluded because they stated that their data should not be used for the analysis, one participant was excluded because of very fast response times to the items e.
Participants were recruited via social networks and an online survey platform on which researchers take part in online surveys in exchange for other people to take part in their surveys. The study was conducted via an online survey platform and participants watched the same video as in the first and second study.
Thus, participants also evaluated the items in regard to an interaction with a technology which potentially evokes creepiness. Afterward, they responded to the CRoSS, the other measures assessing convergent and divergent validity, demographic questions, and similar to Study 1 and 2 to an open-ended question in which they were required to describe what has happened during the situation in the video.
All measures except for extraversion and conscientiousness were rated on a scale from 1 strongly disagree to 7 strongly agree. Privacy concerns were measured with six items adapted to the purpose of this study; taken from Langer et al. Transparency was measured with three items.
Two of these items were taken from Langer et al.
- Agglutination, Complement, Neutralization, and Inhibition.
- Green Planets: Ecology and Science Fiction.
- CREEPINESS | meaning in the Cambridge English Dictionary.
The four controllability items were taken from Langer et al. We adapted these items to the purpose of this study. For conscientiousness and extraversion we used a German measure of the Big Five Inventory by Rammstedt and John with four items for each of the dimensions rated from 1 disagree strongly to 5 agree strongly. Findings showed that they came up with similar explanations to participants in Studies 1 and 2.
- Completely Loved: Recognizing Gods Passionate Pursuit of Us (Loving Jesus Without Limits)!
- Dominic Pettman reviews Creepiness!
- Creepiness: xygylabazo.ga: Adam Kotsko: Books.
- “Cosmopolis” and “Compliance.”.
- EC/EU Fact Book: A Complete Question and Answer Guide.
Regarding convergent validity, Hypotheses 1a—d were all supported as the results showed significant correlations between the creepiness scale and privacy concerns, computer anxiety, transparency, and controllability. As hypothesized, privacy concerns and computer anxiety were positively correlated, whereas controllability and transparency were negatively correlated with creepiness. Furthermore, we found additional support for Hypothesis 1c as the results showed that transparency only correlated with creepy ambiguity, whereas there was no significant correlation between transparency and emotional creepiness.
In contrast to the second part of Hypothesis 1c, there was no difference in the magnitude of correlations between the subdimensions of creepiness and controllability. Regarding divergent validity, the results cf. Neither the entire creepiness scale, nor its subdimensions correlated significantly with extraversion and conscientiousness. To summarize, the results of Study 3 increased our understanding of the construct of creepiness and its nomological network. Taken together, these results provide support for the convergent and divergent validity of the CRoSS and its subscales.
Lastly, Study 3 showed initial support of the assumption by former research that females might express higher feelings of creepiness than males McAndrew and Koehnke, In the experimental design of Study 4, this finding will be investigated more closely, together with the assumption that creepiness is a feeling that can also be expressed in real-life situations. In this last step of our scale development, we applied the CRoSS to a real life-situation.
Throughout the previous three studies, participants only watched a video involving a creepy situation with a technology. However, creepiness should also be present in situations that do not use technology. Therefore, in Study 4, participants were either approached by a male or a female experimenter in a public place where they were asked to respond to the CRoSS items.
This was either done during the day, or at night. McAndrew and Koehnke proposed that men will be evaluated as being creepier than women.watch
A reason for this could be that males are, in general, more physically threatening and underlie the stereotype of being more violent than women McAndrew and Koehnke, On the one hand, this could mean that people are more afraid of men. On the other hand, this also implies that men are perceived as being less predictable and potentially less controllable than women, so other males and females might be constantly aware of a possible threat by males. Additionally, McAndrew and Koehnke proposed that women in general feel more creepiness in most situations.
People who think they are weak might also think that they are less able to control a variety of situations. Hypothesis 2a: A male experimenter will evoke more creepiness than a female experimenter. Furthermore, environmental aspects can also evoke creepiness. For instance, McAndrew and Koehnke describe a dark tunnel as an example of a creepy environment.
In addition, Watt et al. The night relates to our concept of creepiness such that at night people might have the feeling that they are less able to predict what will happen, and that situations that occur at night are less transparent, simply because people cannot perceive their surroundings as well as during the day.
Thus, we propose,. Hypothesis 2c: The experimental situation during the night will evoke more creepiness than during the day. Our experimenters received a script instructing them to dress similarly, to not smile at participants, and to not behave especially friendly, but still politely. In addition, they were told to never collect data at the same time as the other experimenter. The experimenters were both Caucasian, had blue eyes and bright skin, and were 26 years old. Participants were instructed to rate the situation they had just experienced i.
The hours of data collection during the day were between 3 pm and 6 pm, for data collection that took place at night, the hours were from 10 pm until 12 pm. A—F Pictures of the public place where participants were contacted during the day or at night. Copyright Josephine Malsch. In Hypothesis 2a it was assumed that a male experimenter will evoke more creepiness than a female experimenter. Thus Hypothesis 1 was not supported. Means and standard deviations for the combinations of the independent variables gender of the experimenter, gender of the participant, and time of the day.
The current paper introduced the Creepiness of Situation Scale as a measure to examine creepiness of various situations. Following rigorous psychometrical guidelines for scale development by Hinkin , the four current studies show that the CRoSS offers a reliable measure of general creepiness and its two subdimensions emotional creepiness and creepy ambiguity. It therefore offers an additional perspective to evaluate novel technologies over and above scales based on the TAM Venkatesh et al. Furthermore, the CRoSS could be a valuable tool to advance research on creepiness in interpersonal situations.
Study 2 confirmed these two correlated subdimensions of creepiness in a German sample. Finally, Studies 3 and 4 supported the validity of the CRoSS in a technological and in an interpersonal real-world context. Additionally, the results from Study 4 indicated that the CRoSS is sensitive to experimental manipulations based on theoretical assumptions.
As we explained in the introduction, organizations nowadays constantly come up with new services in which algorithms judge human behavior e. However, no study so far has attempted to integrate theoretical assumptions regarding creepiness to develop a sound measure for creepiness. One shortcoming of previous creepiness measures is that they were not developed to fulfill basic psychometrical standards. For creepiness research to evolve however, and to make results from different studies on creepiness comparable, there is need for a psychometrically sound measure of creepiness. For measures lacking theoretical background, it is hard to come up with theoretical assumptions about its relation to other important measures.
It is even harder to develop specific hypotheses. This might be a reason why research has yet to provide validity data on the relations between creepiness and other measures. Regarding validity of the CRoSS, it was possible to generate theory-based hypotheses concerning the relation of creepiness with other relevant measures and to predict the direction of these relations e. All in all, our results regarding reliability and validity suggest that the CRoSS is a potentially useful scale to advance research on creepiness.
An additional contribution of the current set of studies is that the findings suggest that creepiness can be differentiated into two subdimensions, creepy ambiguity and emotional creepiness. This differentiation can help to increase our understanding of the creepiness concept. One example for this increased understanding can be found in Study 3 that found that non-transparent situations evoke creepy ambiguity, but to a lesser extent emotional creepiness.
This indicates that increasing transparency may help to decrease creepy ambiguity. In contrast, influencing situations which involve the affective dimension of creepiness might require other interventions. For example, it is imaginable that, similar to other negative emotional impressions e. These insights would not have been possible with a single-item measure of creepiness. Speaking in favor of the value of the CRoSS for research on creepiness, the current studies support and extend previous research regarding creepiness e. Our studies show that creepiness relates to variables that are associated with the predictability of a situation i.
Study 4 also lends further support for the relation of creepiness and predictability. During the day, it might be more common to interact with people who contact you to fill out some questionnaires, whereas an experimenter who approaches people at night to fill out a questionnaires is rather uncommon. Therefore, participants who realized that an experimenter is approaching them during the night had a harder time predicting what will happen next than participants exposed to the same situation during the day.
Furthermore, our findings support assumptions of Shklovski et al. Studies 1—3 exposed participants to a situation that was interpreted as evoking privacy concerns. At the same time, Study 3 found that participants who perceived the situation as a more severe instance of privacy invasion also reported higher feelings of creepiness.
The Age of Creepiness | The New Yorker
Another field of research that could benefit from the CRoSS is research regarding the uncanny valley. As stated in the theoretical background, this field of research has produced mixed results cf.
We posit that the CRoSS might be a useful tool to explore the uncanny valley in a more standardized fashion. Future studies regarding the uncanny valley may use the CRoSS as an additional and broadly applicable way of measuring its impact, thus making results more comparable. It would be especially interesting to investigate the relation of creepiness and eeriness Ho and MacDorman, in research regarding the uncanny valley. Up to now, we can only speculate that emotional creepiness will be especially related to eeriness. Aside from its usefulness within technological settings, the CRoSS also seems to be a valuable measure for assessing creepiness in other real-life situations.
For instance, we found support for assumptions by McAndrew and Koehnke that women in general express higher feelings of creepiness. This result is similar to findings from previous research which has shown that women tend to report more pronounced affective reactions than men Ashmore, ; Ho et al.
This supports the assumption that creepiness has an affective component. In contrast, our results question the expectation of McAndrew and Koehnke that men evoke higher feelings of creepiness. The results from Study 4 indicate that a female experimenter induced more creepiness, implicating that women possess characteristics e. One possible explanation for this result is that we told our experimenters to be polite, but not to smile. Since females tend to be more emotionally expressive Kring and Gordon, , it might have been more unfamiliar for participants to be contacted by a female experimenter who did not smile as opposed to a non-smiling male experimenter, thus leading to higher feelings of creepiness.
If the reason for more creepiness was unfamiliarity of the situation, this would again speak in favor of the assumption that predictability is related to creepiness. More precisely, people in unfamiliar situations possess less knowledge about the situation and therefore they might be less able to predict what will happen next cf. Therefore, we propose that creepiness might be elicited by situations challenging existing cognitive schemata thus reducing familiarity Rumelhart, For instance, people might possess knowledge and experience about interacting with computers e.
However, it is important to note that the current study only showed initial support that creepiness relates to concepts associated with predictability and familiarity e. The exact paths and causal relations between these concepts and creepiness need to be addressed in future experiments. This series would be enjoyed by fans of Goth Girl and Lemony Snicket as it has a mixture of fun, creepiness and mystery.
It's an chilly urban nightmare whose best moments of skin itching creepiness and nasty violence arrive early. Watch at home. By integrating its Live Context Marketing Engine into media platforms such as AppNexus and MediaMath, Grapeshot seeks to take the creepiness factor out of the audience's online experience.
C-suite conversations. An astute and emotionally charged read, riddled with creepiness. Listening to this excellent performance by the BBC National Orchestra of Wales, conducted by Tecwyn Evans, one also detected a creepiness and a worrying sense of threat to the music. However, according to scientific research about the nature of creepiness , it's far more likely that you'll get creeped out by your taxi driver on the way home from a night out.
The creepiest jobs worldwide. IT'S TIME for fifty shades of creepiness with the return of the hottest serial killer to have ever graced our screens. I feel like we're leaning into the creepiness a little more and the comedy a little more," he said. Lupton is a master of steadily building creepiness and suspense.
The original was a masterpiece of sustained creepiness.
Copyright 2019 - All Right Reserved