Data Science

Emotions and Logic

Kate Dudzik
AI Systems Engineer

Emotions are a critical component in why people, such as our target markets, make the decisions that they do.

It’s one thing to select an audience, define our population parameters, the platforms in which we will reach them – and a completely separate exercise to show prospective customers that this is the right product for them and why.

Often, we think of ourselves as separate from our emotions – under the false belief that our “logical” selves are completely distinct from our emotional selves, or that emotions have no place in logical decision making. We like to believe that we bought a product because it was exactly the right fit for the task, our budget, or our needs - that how we are feeling when we make that type of decision has almost no influence on which product we select. However, this false belief is quite far from our human reality.

The truth is, there is no “off switch” for our emotions or a bridge we cross between our “logic side” and our “emotional side”.

Our cultural preference to consistently acknowledge the powerful, umbrella terms of emotion over nuanced grey areas or less “potent” feeling emotions have a certain degree of influence towards why we seldom think of ourselves as always “feeling” something. The truth is we are always feeling something, which is most of the time, not an extreme emotion. Yet, those intense feelings are the moments that stand out to our brains the most especially when we think of emotions as an experience, and those moments then become stand-out memories we can strongly recall. Most people list the most powerful, sometimes even overwhelming emotions first (and sometimes, exclusively) when asked to list what emotions they experience on the day to day.

Why is this?

Well, from an evolutionary perspective, our survival instincts play a large role in us remembering the bad (almost too well sometimes), and the good so that we try to repeat it. Stay away from the danger and move towards safety and prosperity. Most people are aware of our fight or flight response – the nervous system’s reaction to a perceived threat/danger, and an innate, primal need to either attack it or run. Many people consider this to be an overwhelming emotional sensation, that stands out quite strongly in our memory as something to learn from – and even a situation to replay quite consistently in our minds if we are not being mindful of our thoughts. Those are often in the group of imagined scenarios that like to pop up and replay in our minds, influencing branches of similar “what-if” scenarios.

Furthermore, there seems to be this very strong belief that people are either “book smart” vs. “street smart”; “emotionally intelligent” vs. “intellectually intelligent” in our culture. This isn’t a new trend - people like to put each other in predefined boxes, labelled as one thing or another. In my life, I have often had people try to boil me down to some basic characteristic, as I’m sure you have experienced as well.

Doesn’t feel great, does it? Perhaps it has something to do with the fact that although maybe they are perceiving a more dominant aspect of your personhood based upon their respective beliefs and opinions of you, they are omitting the fact that all humans (including you) are a spectrum and blend of both.

I know as a growing and (hopefully) well-rounded individual, I am both an intelligent person committed to a life of scientific and intellectual learning, as well as someone who is self-aware and committed to emotional growth and development. I can both be a good, sympathetic friend who loves to listen deeply to others, is great at dinner parties, and has a wealth of life experience and someone who can design and develop complex computational systems in both academic and business settings. All of us – yes, all – truly belong in all of those categories; “book smart” vs. “street smart”; “emotionally intelligent” vs. “intellectually intelligent” at different points in between on the scale and at varying degrees of competency, expression, awareness, which is also influenced by our genetics, intellectual capacity, upbringing, culture, values, and environment.

I want to stress as well that not all days are the same – some days, we may feel more prone to what can be perceived as a colder, more distant version of ourselves more in tune with factual stimuli and thoughts - which is a completely different emotional profile than our warmer, more social and self-compassionate days. And truth be told, we are always learning and evolving to strengthen ourselves in different ways of thinking and feeling. This does not negate the existence of the grey area along with both sides as a part of us, at varying degrees, existing at all times even if not being expressed at all times. Emotional experience and expression are highly subject to our perception of the context and situation – what you feel in that moment. This human experience is also highly subjective from person to person, and each of those emotions looks different to/for each person.

There is so much information existing around us literally every second of every day – from the objects and movements within our immediate environments to the blast of social media and news, we are bombarded with a constant flow of stimuli. Our brains evolved to deal with this by applying different heuristics to manage the overflow, directing our attention to what it believes to be important for our goals (more on that in another blog post on heuristics and biases!). Emotions are actually a very important driver in what we choose to pay attention to and guide us to make decisions whether we realize it or not.

When it comes to advertising, the more old-fashioned approach of semantic analysis (referring to positive, negative, and neutral classifications) is useful for high-level filtering of information but is in no way competent when it comes to identifying how the audience is actually feeling when engaging with content. There is such a spectrum in each category of emotion type that ranges in complexity and from situation to situation, context to context. This is why with Reticle, our Emotional Analysis Artificial Intelligence System, we use semantic analysis as a high-level filter to get into the granularity of what was discussed earlier through our Reticle Neighborhoods: we have our powerful, strong, memorable emotional expressions, and we also have more nuanced, normal, daily feelings.

If we are going to reach the right people who could truly benefit from our products at a personal, honest level, we need to be willing to accept the human experience of complex emotions and how they are in everything we do.

With a specific focus on the science of emotion, Kate brings a wealth of experience, overseeing model design, creation and development, sentiment analysis, machine learning, and natural language processing. She is responsible for ensuring the success of our systems and provides strategic guidance to strengthen our media efforts.

More news

Making Ads Matter

In partnership with Peer39, we’ve integrated our emotional classification models into their tech stack, classifying enormous volumes of open market inventory to bring self-serve emotional categories to advertisers, available in Peer39’s Contextual Marketplace.

Supplement Action vs. Replace

No matter how much you plan for and predict, humans and technological systems will always find a way to surprise you. Learn why visualization is a vital component of creation and planning when designing a tool or a system.

The Process of “Debugging” Code

Different projects require different editing processes and strategies to make sure your program is running as it should.