Filmmaking / Directing : Can Science Improving Audience Experience and Returns? by Dr Keith Bound

Well that was an interesting response with no evidence to support your argument. Actually revealing the filmmakers narrative blind spots is important because it reveals where technology and filmmakers assumptions about audience engagement are incorrect. Future filmmakers will integrates science with cinematic storytelling - like I do at this time. As I mention in the press release they are many factors that contribute to poor returns but the 4 year scientific study has evidence based solutions that can improve the way we tell stories without interfering with the filmmakers creative input - instead it it broadens the filmmakers language .

I understand your feelings and agree with you if data scientists were giving instructions to a filmmaker what to change in their story, however I have 27 years in the creative industries, award winning innovator, designer and cinematic storyteller. So I have both skills sets science and the creative so it makes things more interesting because I know what actually will engage the viewer and what wont and I have only scratched the surface. The research identified how fast short camera shots effect disorient the viewer rather than engage them and how different types of story structures, camera frame, lighting, sound (diegetic and non-diegetic) , stage props engage viewers and how acousmatic and ulrasound can increase anxiety in viewers. What the science tells us is that in some circumstances the viewer reacts to a scene differently to what the filmmaker expects. My research/analysis includes 3 data sets, viewer psychophysiological experience, self-reports and a detailed textual analysis of the story structures. Screenwriters and filmmakers will embrace science because movie studios have been doing this type of research since 2010 to understand moviegoers real-time experience. Most of the research has been used in marketing but with interactive experiences, VR and 360 filmmaking, science and machine learning is placing a much more important part of driving viewer experience. I agree some filmmakers will not want to embrace the science but from a business perspective they are more likely to make a films that consistently engage the audience and maximise returns. Moreover it is and will become part of 21st century storytelling. I hope that helps. I have short video to look at if your interested and a short article that will elaborate on the points above

https://www.youtube.com/Resolving Filmmakers Blind Spots

and an article published on here

https://www.stage32.com/blog/Resolving-Filmmakers-Blind-Spots

I agree, Owen, this sounds utterly depressing. More contrivance, analytics, and algorithms. Sigh...

Overthinking.

Beth it seems that you are assuming that it is all down to algorithms which it is not - I am not a computer scientist - if you read what I have said you will see that I am a creative but using science to eliminate contrivance from the filmmaker

Hey Keith Apple computers is hiring. They're gonna conquer Hollywood with their Itunes channel. Original programming. Big money. Take your skills to Silicon Valley. Good Luck!

Me? I'm gonna spend my $$$ on a ticket to see GIRLS TRIP. Nothing fancy or razzle dazzle. Just 4 girls having fun and doing crazy shit.

I didn't mean algorithms literally, but rather figuratively -- sorry, Keith, not clear on my part. To further explain... I find this depressing because as an audience member... when I watch a film I want to see someone's creative vision, their perspective, their creativity, their point of view -- say, the Coen brothers unique storytelling and style, for example. I understand and agree; science/research adds another tool to use for effect, yes, absolutely. But it is concerning when the "artificial" determines visual relevance or "art." That in itself seems to ignore the vision of the artistic individual, her/his personal imprint, how her/his unique creative ability, human experience, personality, intelligence and artistic skill imbue a piece of work. ;)

Wow! thanks everyone for you amazing input and feedback. Like I said in the press release I mention in the blog there are many reasons why films flop at the box office - however that are also one that do impact engagement as found in my 4 years study. What is interesting as Sarah points out film is not longer confined to the cinema instead film is used in many types of media experiences - such a games, transmedia experiences and more of us are watching films on tablets. Although the largest audience by age groups 14-25 has declined over recent years this group have stopped going to the cinema because the right content is not shown - well that what researchers have found. Owne interesting point about as you put it unemotional audience responses in stories. Based on my research across genres there is always some form of story conflict/crisis even in comedy - this is quite crucial to measure audience engagement so although the research did not cover all genres (another research project) we have a starting point to understand audience engagement in real time - Thanks Dan I will look into Apple - appreciate your comments too

Beth thanks for you very comment the process I have developed does not interfere with the creativity of cinematic storytelling - remember I am a creative too! to a greater understanding of what works when engaging an audience - VFX is no different because it has a technology side to aid creativity - the science of storytelling does the same thing

You're welcome, Keith! Give me a job at apple! The employees eat for free at their compound! Also Jeffrey Kaztenberg is doing 10-minute cellphone programming for 18-35 age group.

Hook up with JK and you can buy yourself a mansion, lecture TEDs Talk world-wide, be the point of the spear ! Talk your expertise to China. They believe hype :))))) Happy hunting!

Hi Owen - I love your enthusiasm - I am not slipping past anyone LOL - you laugh at my term the science of storytelling yet many filmmakers use some form of psychology to create their stories. I suggest you do some research I can send you some great links if your interested -there is a lot of research by cognitive psychologists and neuroscientists in how we understand stories, film and audience experience. What I propose is giving the filmmaker many options to allow their creativity to flow and not just one option decided by data.

Thanks Dan - I will have a look into JK - Thanks - if I get a job at apple I will contact you - I never forget people who help me on my amazing journey

Owen you are missing the point - I am a filmmaker and I don't have a white coat - creativity is abound I don't use algorithms - I think you're suffering from cognitive dissonance

I am a man of science or atleast I comprehend 70 percent but the rest is dark matter to me. As for the science of storytelling, I am not convinced. I need the data that supports the correlation between sweat, emotions, and interest. Do you have data or do you need artist to participate in a study? What scholarly journal is your research published?

Dr. Bound, I am giving you a hard time but you get my point. If we use words like science and data then the burden to provide it is on you. I have to admit, you are the first entertainment scientist that I know. That makes you special. Congrats on your research. I hope it is okay to message you and discuss this further. What can I say professor, all this sweat on me must means I found this engaging.

Owen unlike you mine is NOT an opinion My 4 year study (PhD) research has been independently assessed through a viva by two world leading professors - I passed my PhD without corrections placing me in the top 1% of PhD's in the UK. I have presented my research at 6 peer reviewed international conferences in USA and Europe 2013-2017 - take a look at my website www.recptivecinema.com Before you criticize peoples research and work based your opinion (blind spot) that I could get from any layman please have the decency to research my work/research first.

Myron DeBose no your not giving me a hard time see my previous answer to Owen about my PhD - yes sure we can talk more here are some links for you www.receptivecinema.com and https://www.youtube.com/Resolving Filmmakers Blind Spots and an article published on here https://www.stage32.com/blog/Resolving-Filmmakers-Blind-Spots send us a message at keithbound@receptivecinema.com Best keith

Keith, I have greater respect for you because you did not bring up your super nerd PH.D. Plus, I am not insecure. Was your research used during this past American Presidential election? I saw a live stream of that graph image on CNN, MSNBC..etc. I do find institutional research interesting. Expect push back but a positive attitude will cause open-minded people to give your info a chance. Nice meeting you Dr. Bound and Owen comments helped me notice your post. It takes all kinds to help get our message out.

There are not many people in the world who really have to tell a story. They don´t care about blind spots, they also don´t care about what the filmindustry like.

Hey keith

send me an Ipad when you're working for Apple! :) Good Luck!

I recognize Keith's budding Guru status. No one has to respect the work of Academia but some understand its importance to industry and society overall.

Owen you don have the slightest idea of the work involved and intelligence required and the rigors of the research to complete PhD. I dont need to justify my research to anyone as I did that in the viva - hence why I passed my PhD and in the top 1% of PhD's in the UK. However, I am pleased you are reading my 100000 word thesis - enjoy you will actually learn something , especially my contributions to knowledge.! However, my research has also moved on from my PhD and I am working with television and film companies.

Owen I thought you said you were reading my thesis and my contributions to knowledge in films studies, media psychology and psychophysiology. if not please keep your assumptions to yourself

Owen I am already working with content providers on a global basis who collaborate with me based on my research findings and methodology - they see it as cutting edge and adds value, so your laymans opinions and assumptions are of little consequence - although I have enjoyed the banter with you I dont really have the time to indulge you further

In reading the material posted here and in the blog, Keith’s research is fascinating. I have a bunch of questions that could probably be answered by reading more about the results of the work Keith has done. Knowing how people are actually reacting to what I do seems unquestionably to be a good idea.

Keith where can I find out more about the results of your research?

Thanks Andrew for your enthusiasm and comments although you can find out more about my results you would need to access my thesis through the University of Nottingham website. Remember that the research results focus on specific narratives structures e.g various, shared, direct and composite suspense. However, since my PhD I have developed my research further over the last 12 months and are now working with content providers, due to intellectual property issues, novelty, contractual issues I can't discuss these aspects of my research. Hence why Own was referring to my results being vague. However, if you would like to discuss some of my PhD results through a short skype call email me at keithbound@receptivecinema.com thanks again for your interest.

I will read your thesis, should be interesting. For now just a couple of questions. I understand your methodology as described and why you chose clips from horror films. Do you have any information on cultural differences around the world having different responses? Do you have any information on how accurately your research applies to other genre's, i.e. Romantic Comedies?

Andrew, good questions. Although there are cultural differences concerning story context, with hollywood narratives being popular in the west and now it seems in the east (notice Hollywood movies studios are tapping into the Chinese market , especially paramount pictures because some blockbusters that did not do so well in the USA but have been successful in China. e.g one of the Transformers films I think released in 2015). China's film industry growth is around 39% compared to the USA at roughly 10% and China is expected to be global market leader by 2020.

In psychophysiology, electrodermal activity with records electrical changes in the skin (emotional sweating) there is little evidence to suggest that different races respond differently. When we consider that Hollywood films have some form of story conflict/crisis across genres, which often produces suspense (mediation of a threat to character or audience) creating an anticipatory stress response. The conflict triggers our body's defense system (fight-flight response) and the Amygdala triggers a neuro-network to enable our body to respond appropriately to the threat (real or vicarious) through the sympathetic nervous system, (a branch of autonomic nervous system) which creates tiny electrical changes and emotional sweating. This process works across cultures because we all have the same body defence system. In my study I was able to measure anxiety in terms of durability and intensity to people watching specific generic narrative structures from horror films. However because conflict/crisis cuts across genres you should still be able to measure very intense anxiety (dominant emotion) responses from a romantic comedy. Moreover the novel techniques I use to measure anxiety in terms of durability and intensity could be applied to other dominant emotions. (bearing in mind we often experience a range of emotions at anytime e.g you find out your partner have been unfaithful - you could feel love, jealousy, anger at the same time). I am now working on measuring many other narrative features and cinematic techniques based on 'cause and effect' paradigms. I hope this answers your questions.

Owen yes they maybe difference between cultural, sex and possibly age differences but this has not been proved was NOT the object of the study. Electrodermal activity is one if not the oldest psychophysiological recording technique (100 years old). In my study I had a mixture of people who loved horror films, e.g. students from malaysia, people who worked in different occupations and the sex was roughly 60-40 male - but again neither of these factors were the focus of the study. A PhD is like most research projects is not about finding everything and all possibilities but to focus on a specific research question(s). My research project was about defining suspense because there is no concise definition of the experience of suspense and whether it is quantifiable. Not only did I define suspense but also created a novel methodology to measure the experience of suspense which contributed knowledge to film studies, media psychology and psychophysiology. All the big Movie studios , video games companies, consumer research organisations (e.g Nelisen) use neuroscience (brain imaging) electrodermal activity, Eye Tracking and many more have been using psychophysiological recording techniques since 2009 to help them understand viewers, users and consumer behaviour.

To give an very brief example of identifying narrative blind spots. Silent House - direct suspense film clip. The director constructs suspense through the eyes of a female protagonist (he polaroid camera) a sequence of 3 still images and finally a moving image with a man running towards the her to attack her (it looks like he is attacking the veiwer. In between the bright camera flashes, still images and moving image there are periods of complete darkness and the viewer can only determine what is happening through. the diegetic acousmatic sounds. All the still images and moving image are a dutch tilt which is assumed to create anxiety in the viewer because of the disorientation of the image. Okay to the viewer first watches a camera flash and an empty room, then the screen goes black (longest section of darkness) the viewer can hear a female whimpering in the background with a change in pitch. The 2nd image is a young girl in a hallway the protagonist screams loud, then the screen goes dark again, shortly later another still image of an empty room, then another short period of darkness followed by the guy running towards the viewer to attack the protagonist) then it goes dark again. Okay so what did I find, well this sequence did not produce the desired viewer response even though studies have shown that we engage more strongly when we watch a moving image (at the end) I detected the strongest anxiety responses in terms of durability and intensity during the long period of darkness after the first image, but when the 2nd image of the young girl appeared anxiety was reduced and the same with image 3 and the moving image at the end. The reasons why viewers said they were more anxious when the screen was black but because they did not see the protagonist and 86.6% of the film clip was a dark screen they found it difficult to know why she was so upset. Other viewers found it predictable and guessed the pattern of suspense while some other viewers expected the images to be scarier. What was interesting is that during the prolong anxiety response during the longest section of darkness wast the protagonists whimpering sound which changed in pitch - anthropologists have found that animal cries similar with a shift in pitch produce stronger anxiety responses that a sudden loud piercing scream which was the case in this film clip. So this is a brief example of the narrative blind spots - e.g viewers need sufficient visual information to understand what is happening to a protagonists - sound alone works for a while but then fades off. Also the study also found how camera shot of the protagonist to empathise with the characters feeling of terror ie.g ECU that reveals the protagonists facial expressions of fear - this is also backed up by film theorists and science in how we empathise with other people in pain (unless we have psychopathic tendencies). So one idea that could have produced a stronger anxiety response would have been the 2nd image of the protagonist screaming - e.g she drops the camera (changes to moving image) and as it flashes the viewer can see her in ECU and her facial expressions as she screams with the man shown walking towards her behind her in the same image - then the image cuts to the dark screen again. This is one example but like I said the whole suspense pattern could be altered and make the dark screen sections shorter...hope this helps to give a brief example. If I explain this fully as in my thesis it would take at least 1-2 hours..there is a lot of detail missing..but it should give you an idea

The cultural differences I was thinking about were more along the lines of differing cultural norms than racial differences. One simple example is wearing black or white at a funeral, which is culturally correct? Same goes for weddings. The choice varies around the world. Also there are differences in story telling patterns. Nice to know that there were no differences between races in your research response showing yet again people are people.

I’m guessing that the intensity of the reaction you are monitoring would be less in a Rom-Com than in a horror picture. Assuming that you can measure differences in the response level (haven’t read your thesis yet), might that the case in the Horror/Rom-Com comparison? Also would the intensity of the reaction be greater with a cut to a shocking scene from an emotionally neutral scene or from a scene that is building the tension? Is it equal? If in measuring suspense, if the neutral scene got a 0 and the cut to the shock scene raised the reaction to a 1, would a suspense scene that is at a 1 be raised to a 2 on a cut to a shock scene? Is the additional tension from the cut equal regardless of the starting level of tension.

You used clips from pictures to gather your data but is there a difference in the reactions you studied over the duration of a full length movie? Is the reaction to the clip different when it is seen by itself as opposed to when it is seen in the context of the longer movie? If the scene you used as a clip came after an hour containing a series of suspenseful scenes, would the reaction be greater or is it lessened as the audience perhaps adapts to the prolonged tension?

Thanks for the information, good to hear that your research is continuing and expanding into other areas.

Thanks for your questions. In regard, to the different cultural aspects you mention it would be difficult to measure and I am only interested in Hollywood style of storytelling because at presents the most popular film in the west. It’s a case of one at a time. Like I explained to Owen we can never measure everything. To identify a research project you have to do a lot of research by investigating the academic literature and design your experiment which is often far more complex than most people think – I went through a massive learning curve during my study.

You make a good point that measuring Rom-Com might be different in terms of measuring an anxiety response during a suspenseful scene (e.g The first Ted film he was kidnapped and left to be tortured by that nasty child – then he escaped and big chase, but filmmaker tend to use similar conventions to build suspense as in horror films – I have not tested ROM-Com but even a romance film when lovers are separated for some reason (conflict) could make some viewers just as anxious as if it was a horror film if they have empathised with the characters. In my introduction to my thesis I do talk a lot about the cultural aspects of suspense. For example, when filmmaking began feature films could not be produced, instead the audience watched short episodes (serial filmmaking) using the cliff hanger so that audiences turned up for the next episode. They even promoted suspense across genres because suspense would hook the audience no matter what genre – Perils of Pauline.

To measure suspense as an anxiety response I used some generic tiny narrative structures from film scholars Susan Smith’s model of suspense. You see a lot of neuroscientists and some companies will just measure viewer responses to the duration of the whole films. At first this seems logical but you cannot attempt this when you don’t know how viewers respond to ‘cause-effect’ paradigms (narrative/cinematic techniques) that can be measured. Cinematic storytelling has so many different layers, narrative structures, cinematography, sound, mise-en-scene etc.

Therefore, my study investigated tiny narrative structures: vicarious, shared, direct and composite suspense which were extracted from 8 horror films so we can learn more about the complexities of analysing film in small chunks. The results provide a break though in how we can measure cause and effect paradigms – a scene, shot, whole film and the cause-and effect regarding the viewers’ response to camera frame, shot type, lighting, diegetic and no-diegetic sound, stage props etc. For the last 12 months I have built on my research which is leading to more interesting findings regarding to feature films.

thanks for the in depth response Keith. Yes, I have to read your thesis. Actually looking forward to doing so. Not having read it is my excuse for asking about things you've not studied yet. Promise, no more questions until I read the thesis.

You can have all the science, research, and analytics, at the end of the day there's no science to determine who will buy a ticket and put their butts in a seat. When I was at a major studio there was every indication a film releasing that week was going to be a hit. Even the president said, "It looks like we have a hit!" It died a quick death at the box office that weekend. The following Monday the president said, "I don't understand what happened." I do. No one wanted to see the movie even though all the analytics said, "Hit!".

Lindbergh the scientific approach Keith is applying is unlike anything else I have heard of. I’m interested because it might make me re-think story telling methods that have just become a given from decades of using them. A fresh approach based upon unique measurable data seems worth looking into. If Keith’s research can help us to elicit the emotions we want from the audience in a more effective manner it is something worth learning more about. Information has it’s own intrinsic value, but how you choose to utilize information or not is a choice that should be better if it is made with the full knowledge of the value of the effect on the audience, not having the effect merely be a happy coincidence.

A.S. ahhhh..... this is Filmmaking / Directing. If you don't think it has any value then you don't have to bother with it.

You are most well Andrew and you are correct in that my scientific process is very different to the movie studios. Lindbergh Movie studios do use psychophysiological recording but the data is used to promote films (movie trailers) rather than use the data to support filmmaking creative decisions. Also neuroscientists create a specific audience engagement curve but they dont apply this to specific narrative features and cinematic techniques because they dont have an in-depth understanding of storytelling and would not know what narrative features/cinematic techniques to improve engagement. Instead they inform the filmmaker when viewers where disengaged. This leaves the filmmaker in a dilemma because what he/she would not know what to change to improve engagement - hence the filmmakers' blind spot. My research offers an opportunity to resolve these blind spots.