May 10, 2023
WeVu: Using AI, Psychology, and Classical Rhetoric to Find Something New to Watch
On the misery of finding something new to watch
If you’re like most of us, this scenario is familiar: It’s Friday night and time for the two of you to unwind with a movie. So you open up Netflix and then it hits: The dread associated with finding something new and different to watch. You check the newest releases and nothing quite grabs your focus. You then scroll to the recommendations and see a few things but again, nothing quite compels your attention. Your watch list is a few years old, so you start scrolling down, then across, then down, then across….
“How about that one?” “Or that one?”
“We watched that last month.”
“Okay, then how about that one?”
“Maybe.”
So you keep scrolling. And you both start losing confidence in your ability to find something new and different. Twenty minutes later, you’re re-watching Oceans 11 for the 35th time.
Why is it so difficult to find something new?
It is fair to say that we live in the golden age of TV. To satisfy the more than one billion people around the globe who subscribe to streaming services, they produce an astounding amount of content to satisfy the tastes of any and all audiences and keep those monthly subscription fees coming in. More content, more choice. Which is great, right?
Analysts estimate that at any given time, there are 36,000 hours of content in over 5,000 titles on Netflix. Hulu features an estimated 40,000 hours of on-demand content any given month. And Prime Video, the beast of category, is estimated to feature an astounding 28,000 movies and shows available on demand at any given time. So why is it that finding something to watch, especially together, has never been more frustrating?
Because there are so many choices, our streaming services must filter among movies and shows, so they use recommendation algorithms to both categorize and recommend them. Without some way of categorizing the overwhelming amount of content available, finding titles without some form of recommendation would be an even more dreadful experience.
The simplest way to make recommendations is to pull from the library movies and shows that share similarities with movies and shows you’ve already watched. But here’s the rub: What you’ve watched before is not necessarily an accurate reflection of what you really like. And that’s why you watch Ocean’s 11 for the 35th time instead of finding the gem among the thousands of titles available to you.
The science behind the way we choose
Scientists studying how people make choices have discovered a paradox in the way we evaluate options. While we like to believe that more choice means more freedom and more agency, more choices burden our brains with a cognitive burn-out.
Further, our brains must balance reason and emotion when making choices. Behavioral scientists have found that when people lose the brain function associated with emotion, they are unable to make choices at all.
In How We Decide, Jonah Lehrer’s fascinating 2009 book about decision modeling and cognitive neuroscience, the author argues that our ability to make decisions requires a balance reason and emotion. When people lose the capacity for emotional consideration in decison making, they’re simply unable to make a decision. They get caught up in a cycle of rationalizations that return equal consideration for each choice. When people lose the rational balance in decision making, they choose everything all at once, an equally debilitating outcome.
Further, a research study conducted in 2018 and published in the academic journal, Nature, showed that there is a similar debilitating effect when the decision involves too many choices. But what makes for too many choices? The answer depends on the value people place in the resulting choice.
For low-consideration decisions, such as selecting apples from the market, the number representing “too many” can be in the dozens and dozens. When decisions are more valuable, such as finding a movie to watch, the number representing “too manuy” may be 10 or 20. The mental effort to consider each choice eventually overwhelms our cerebral cortex and we simply remove ourselves from the act of decision making. This is when we watch Ocean’s 11 for the 35th time.
Researchers gave participants varying sets of items to choose from. While the subjects made their choice, the researchers conducted MRIs to gauge brain activity. The results showed that too few choices in the set provoked a minimum of brain activity, leading them to conclude that the set didn’t contain enough value. Conversely, when choice sets included more than 24 items, MRI scans showed a sharp uptick in neural activity, followed by an equally sharp decline. Participants just gave up making the choice when the set was overwhelming.
In a world of endless entertainment options, abundance can be the enemy.
But what if there was a tool that helped us find something new and overlooked? Something that brings new content to light and solves the problem of finding something new and different but not too different? That’s WeVu, a new recommendation tool that will make finding something new to watch easy.
Three Things WeVu Does Differently
- WeVu gets to know your personality
- WeVu only makes a few recommendations
- WeVu lets you pair your recommendations with another
WeVu gets to know you
The problem WeVu solves is finding great content recommendations for the next movie night. To do so, WeVu features a dynamic quiz that gets to know your personality based on the choices you make in the quiz. The psychology of decision making shows a quantitative relationship between choice and personality. The quiz allows WeVu to power your recommendations based on who you are, not what you’ve previously watched.
In information design, this concept is known as a behavioral recommendation. However, we frequently watch movies and shows again because the pain in the decision process pushes us into cognitive deficit. We fall victim to decision fatigue, further reinforcing the behavioral recommendations and further removing us from discovering content we maytruly like.
And that’s why the recommendations you see on your streaming services are frequently so useless. They attribute any choice you make as one you intended rather than one you reluctantly made just to get something started. And this is precisely why Netflix launched the “Play Something” feature: To provide a failsafe against closing the app. It is the behavioral recommendation of Google’s “Feeling Lucky” search option.
WeVu doesn’t overwhelm you with recommendations
Launching our streaming services can be a bit daunting: We immediately see dozens of choices. Whether recommendations based on previous viewing, announcements for new content, or most viewed content, the launch screen presents a choice set in the dozens. And based upon the research noted above, too many choices immediately challenges our cognitive capacity.
Conversely, too few recommendations does not necessarily make for easier decision making. If our brains suggest to us that the choice set is too small, we discount all individual choices in the set regardless of whether or not the best choice may be included.
So the challenge in making recommendations is to find the right number of choices in the set. That’s why WeVu presents (6) movies and/or shows that our algorithm finds in line with your aspirational preferences. And the UI highlights the strongest match to allow your brain to further reduce choice making.
WeVu lets you pair with another registered user
What really sets WeVu apart is the ability to pair recommendations with another person. Registered users can invite anyone by email to join WeVu. After registering, WeVu users can submit a pair request that to another user that, once accepted, indexes your collective recommendations.
This means that WeVu matches movies and shows that fit within your complementary tastes, showing only movies and shows you’ll both like. This means that the anxiety in finding something to watch with another person goes away. We think this feature is the most critical differentiating feature. While it’s great to find new content when viewing along, it’s significantly better when you can discover new stories with other people.
How WeVu Works
WeVu is crafted to combine the uniquely human with the scale of AI. Both the algorithm and data model apply both rhetoric and psychology to
- The quiz
WeVu gets to know you through a simple quiz. In it, new users select the movie characters they like most from among a set of three to four. This question repeats three to five times, allowing the algorithm to assign the user to a personality archetype. The last question in the quiz allows users to indicate the kinds of stories they find most appealing. This teaches WeVu to scan data for both personality and story type markers.
- The data
WeVu data model is scrupulously designed to categorize movies and shows on dimensions much more human than simply “Action & Adventure” or “Thriller.” Instead, the data model elevates key dimensions such as the story type and the personality archetypes of its leads. Even further, WeVu adds emotional dimensions to the movie that allow it to look beyond typical taxonomies found on virtually every streaming service launch screen. That’s because we know, as does WeVu, that great stories and great performers frequently move beyond a one or two-word description.
- The filters
WeVu then allows users to filter among recommendations, knowing that our interests are not always the same when we’re looking for something to watch. Our algorithm tags movies by the mood of its audience, so you can select from up to five mood types. You can also filter among movies and tv shows, allowing you to isolate each when you’re leaning one direction or another. We’re also working on filtering by streaming service so you can focus recommendations on only the platforms available to you.
What’s next
We’re engaged in a series of conversations to find the right partners. Our priorities are growing our AI-driven content tagging feature and finding the right backers to help us realize the opportunity. If you’d like to know more, visit us at WeVuapp.com
