How to Be More Self-Aware by Letting Artificial Intelligence Look at Your Journal Entries.

Regular journaling helps you become more self-aware. But with apps that use artificial intelligence, you can measure your mental health much better.

When you walk into a doctor’s office you are not supposed to know your temperature, blood pressure, or heart rate. They just measure it. And how do you measure your mental health?

Well, you know. Just record how you feel at any given time or day, track this, analyze the situations that caused you to feel distressed or happy, and develop healthier habits.

Keeping track of your mood and emotions indeed improves your mental wellbeing. But there is one thing. We aren’t that good at understanding our emotions. Just like we aren’t that good at figuring out what exactly makes us physically ill unless we have a degree in medicine.

If something is hard to figure out, it must have a scientific term. Human emotional intelligence (or EQ) is the word we’re looking for. EQ is the ability to recognize emotions (our own and those of other people).

If you’re lucky and have a higher emotional intelligence than the rest of us, you are bound to be successful in your professional and personal life. High EQ makes you more likable, persuasive, a better leader, and a healthier human.

Unfortunately, not many of us can intuitively differentiate between emotions. But this doesn’t mean we shouldn’t try.

Also, there is an alternative solution out there. Technology can help us out! As always.
What is Emotion AI?
Our human intelligence – including our ability to understand how we feel – can be simulated by machines. This is something people call artificial intelligence.  

AI is a very broad term. It encompasses many things and has a variety of applications. For example, you’re dealing with AI when you’re listening to recommended music tracks based on your interests on Spotify or Apple Music. Your Siri is an AI-based speech recognition system. When you’re searching for information on Google, you get relevant links thanks to a sophisticated search algorithm based on AI.

There is a special branch of artificial intelligence called Emotion AI that takes care of quantifying your mental health for you. It is based on natural language processing technology, a big part of artificial intelligence.

Fitness bracelets do a great job at tracking your sleep, monitoring your heart, and calculating the number of steps you take in a day. Apps powered by Emotion AI are like fitness bracelets. They track your mental state on a daily basis.

But how does AI track your emotions? No, it’s not about embedding a chip into your brain. Not yet, at least.

One of the oldest techniques associated with emotion tracking is journaling. What we write can tell a lot about who we are.
What our writing tells about us
The words we use don’t just convey the information. They reflect who we are, how we feel, and how we think. This is proved by several decades of research in the area of psycholinguistics.

Let me show you one example of these researches.

James W. Pennebaker, a social psychologist, spent a great deal of time studying expressive writing as a way to help people get over traumatic experiences.

Trying to understand how what you write influences your mental state, he looked at hundreds of essays written by people who took part in his experiments.

He found out that so-called content words that convey a certain meaning don’t matter as much as function words such as pronouns, articles, auxiliary verbs, and prepositions. It turns out these little parts of speech can tell a lot about what we think about ourselves, our relationships with others, and our attitude to topics we write about.

For example, people who use third person pronouns such as He, She, and They are much more socially engaged and care more about other people than individuals who don’t use these pronouns as much.

Another interesting observation is the higher your status, the less possessive pronouns such as I, Me, and My you use. The lower your status, the more often you’re to use I. This is because higher power people pay more attention to the world, while people in a lower status tend to be looking more inwardly.

Also, depressed people pay more attention to themselves when they’re in pain. They use I very frequently and tend to be extremely honest in their writing.

And how do you think liars talk? They hold off and distance themselves, so don’t expect to see many I’s in their writing and speech.

Check out this brilliant video where Pennebaker explains the results of his research in greater detail.

Reading psycholinguistic studies sounds like an exciting way to spend your time! If you want more of them here is a link for research references – the science behind IBM Watson’s Tone Analyzer service.

Speaking of IBM, it has done a lot of work in the field of AI and emotions. If you heard about Watson – the supercomputer – you should be aware. So how do AI services like IBM Watson detect emotions in text accurately and automatically?
How does AI understand your emotions?
IBM uses AI and machine learning (the most common application of AI), to access your writing data and learn from it automatically, without human assistance or intervention.

These tools have been trained on vast data sets and validated by actual humans to recognize patterns in the language. They can detect joy or sadness, fear or anger even if you say that you’re totally fine. To find your real emotion, Watson studies your whole sentences (important for context) rather than looking at individual words you write.

To detect emotions in text Watson uses complex algorithms. Here is a basic explanation of how these algorithms work: firstly, the system needs to recognize a sentence and assign a syntactic structure to it. This is called syntactic parsing. Then an algorithm does the semantic analysis: it analyses the context surrounding a word or a phrase to understand its meaning. And from there a system finds specific words and assigns them to a cluster of emotions.

At this point, Watson’s Emotion Analysis API detects five emotions – Joy, Fear, Sadness, Disgust, and Anger. It benchmarks emotional tone categories against standard emotion data sets such as ISEAR and SEMEVAL.

The process I described above sounds pretty simple, but in reality, it’s far more complex than that and has a lot of nuances. For example, a word “terrific” can reflect both Joy and Fear. We can only understand the emotion if we know the context. And this is why you need machine learning. It will make the right decision given it has been properly trained on a large amount of data.

Read more about the science behind IBM Watson’s Tone Analyser service here.
What journaling apps use Emotion AI?
There is a handful of apps for tracking emotions. They include meditation apps and various mood trackers. These apps help you record your mood and emotions, but figuring out how you feel is totally on you.

But are there actually any apps on the market that integrate the good old journaling practice with the cutting edge technology?

The truth is, there aren’t many. But I did find a couple that use artificial intelligence to detect emotions in text.

One of these apps is Reflectly. It uses AI to talk about your day and capture your emotions and state of mind. It also gives tips to improve your mood. Reflectly is a good choice for those who don’t write long journals and are more interested in reflecting on their moods and feelings throughout the day.

If you like writing long journals, there is an intelligent app for you too. It’s called MorningPages. The name of the app might sound familiar to those of you who read a book by Julia Cameron “The Artist’s Way.” In this book, the author introduces morning pages – the daily practice of stream of consciousness writing. The app is based on this idea.

There are two things you can do with the MorningPages app: keep a journal and keep track of your mental state. The app uses IBM Watson’s Tone Analyzer service to power real-time emotional analysis. Watson runs its algorithms on your journal entries and tells you what mood you’re in, what emotions your text expresses, and what topics you write about.  

Because you’re naturally curious to see what the app is going to tell you every time you make an entry, this feature in the app will motivate you to journal regularly.

To conclude

Understanding your feelings leads to improving your emotional wellbeing. Of course, you can track your mood yourself, without fancy AI tools. But we believe that Emotion AI helps people become more aware of their emotions. And with the tools like Watson being developed to assist developers, expect to see more AI apps that help people improve their happiness, confidence, and productivity.

Comments