Pattern Recognition

I’ve had many ideas revolving around this concept of ‘pattern recognition’ running through my head lately, so I figured that I’d start putting them into words in hopes that I can form a coherent idea or two.

[Introduction]

What do I mean by pattern recognition?  It’s very simple to understand by example.  Lets say I give you the sequence of numbers 1, 3, 5, 7, … and ask you to provide me the next number in the sequence.  Most people would say that 9 is the next number in the sequence.  This is because they recognized the pattern of me listing off the odd numbers.  This is what I mean by pattern recognition at the most basic level, but as you’ll see as I go on, I believe the concept extends to quite a lot of things…

I apologize in advance for the amount of over-simplification and hand-wavy arguments I’m about to make.  I’m actually intentionally avoiding quoting scientific studies in this blog post, as I’m just treating this as a thought experiment.  Perhaps I will do a follow up on this blog post in the future that will test the validity of any conclusions that I draw from this thought experiment.

[Intelligence]

Doing a quick Google search reveals that many questions used to measure Intelligence Quotients (IQs) can be ‘boiled down’ to testing a person’s ability to recognize patterns.  For example, looking at this site, they provide a few examples such as:

• Which number should come next in the pattern? – This is clearly based on pattern recognition, as evidenced by the fact that the word ‘pattern’ appears in the question…
• Find the answer that best completes the analogy: – This is less directly based on pattern recognition, but you can see that it’s based on the ability to recognize a similarity or relationship between words.  In their example “Book is to Reading as Fork is to …” you’re intended to recognize that you read with a book.  Then you ask yourself “well, what do I do with a fork?” and you arrive at your answer because you recognized the relationship between the words Book and Reading.  This seems to boil down to pattern recognition as well…
• Which of the following can be arranged into a 5-letter English word?
  • This one, once again, is not directly based on pattern recognition.  It is, however, based on your ability to recognize arrangements of letters based on things you’ve seen in the past.  For example – seeing ‘H R G S T’ and ‘W Q R G S’ both immediately throw of “red flags” for me just because neither has vowels.  Then the second one has a ‘Q’ but no ‘U’, and I don’t know many words that have that property.  Based on that, one could pretty easily decide to discard those two options right off the bat.
  • The remaining arrangements of letters ‘R I L S A‘ and ‘T O O M T‘ would presumably be tested by going through random iterations in your head until something clicked and you recognized a similarity/relationship between the letters.  Like maybe for the first one you see ‘S A‘ and ‘I L‘ and think ‘S A I L‘ must be part of the answer so you arrange mentally it as ‘S A I L R‘.  Then you immediately recognize you can swap ‘R‘ and ‘S‘ and you have a legitimate word ‘R A I L S‘.  This once again seems to boil down to some form of pattern recognition, albeit less direct…

My conclusion here is that our idea of general ‘intelligence’ (measured by IQ) as humans seems to be heavily correlated with a person’s ability to recognize patterns, similarities, and relationships.  It’s not clear which way the implication goes (or if it matters).  Perhaps one of these is correct:

• People are good at pattern recognition because they are intelligent.
• People are intelligent because they are good at pattern recognition.
• Intelligence and pattern recognition are one in the same.

It’s very easy to think of cases where people are considered to be intelligent because they saw a pattern emerging that no one else did (or at least no one else took action on):

• Steve Jobs saw the market potential for Personal Computers before many others did
• Some folks (like Michael J. Burry) saw that the sub-prime mortgage crisis was going to occur in 2007 before many others did, and shorted the housing market in anticipation.
• Warren Buffet is often attributed with the ability to see patterns in the stock markets that no one else can, which explains his statistically improbable success in the stock market

[Humor]

Some (but not all…) humor seems to be based on what I am calling pattern recognition.  I’ll try to explain a few examples here, but as I elaborate more it seems that it gets more difficult for me to put my thoughts into words:

• Sarcasm – This is usually based on the ability for you to recognize a deviation in a person’s normal behavior.
  • For example, I overhead a conversation where someone (Person #1) was talking about traveling to a local place that kind of sounded like ‘Canada‘ to get a drink.  Person #2 found this strange (since Canada isn’t particularly close by) and asked “You’re going to go all the way to Canada this afternoon just to get a drink?!“.  Person #1 replied back (sarcastically) “Yes, that’s exactly what I said.  I’m going to Canada.” to which both of them got a good laugh.
  • In order to “recognize” the humor in the joke, one would have to have some context (in particular, the knowledge that Canada is far away, and that Person #1 doesn’t make it a habit to go there with any frequency).  At a more basic level, this is based on the ability to recognize the pattern (e.g.: When Person #1 goes to drink, they tend to go to local bars.  This is a pattern.  Deviating from that and going to Canada breaks the pattern, and it seems extremely unlikely.)
• Irony – Looking at the Wikipedia article on Irony, one of the examples provided is “In the Dred Scott v. Sandford ruling in 1856, the United States Supreme Court held that the Fifth Amendment barred any law that would deprive a slaveholder of his property, such as his slaves, upon the incidence of migration into free territory. So, in a sense, the Supreme Court used the Bill of Rights to deny rights to slaves.”
  • In order to “recognize” the humor in this, one would have to be able to see the similarity (in that both the slaveholder and the slave are humans, who presumably have rights) and then recognize the corresponding dissimilarity (in that the slaveholder was being provided rights that the slave wasn’t) despite them both being humans with the same “Bill of Rights”.  It’s funny (at least in my mind) just due to the absurdity, but the ability to recognize the humor is once again based on a person’s ability to see recognize similarities / dissimilarities.
• Parodic / Hyperbolic – One type of humor that my friend and I find funny is taking something that we dislike, but to the extreme.  For example, in general – my friend and I find people who are overly-patriotic to be somewhat annoying (although honestly, I don’t care that much).  As a result, when my friend says things to me like “God bless America” and “Just doing my job and protecting our Freedom” – I know that he’s intentionally being over the top.  The reason behind this (as you may have guessed) is that I recognized the pattern of his past actions, and recognize that when he’s saying overtly patriotic things to me that it deviates significantly from his “norm”.  I also may have recognized the tone in his voice, and that whenever he uses that particular tone that it’s generally associated with something intended as humor… another “pattern”.  This type of humor can be made even more interesting/fun when there are unfamiliar people on the receiving end of the parodic / hyperbolic statements, or when there are no tones present to provide that “guarantee” that they’re intentionally being over the top (e.g.: written communication)…
• Anti-Jokes – Looking on the ol’ Googley, you can find many examples of anti-jokes.  I’ll quote one here as an example: “A horse walked into a bar.  Several people got up and left as they spotted the potential danger in the situation”.  In order to ‘get’ this joke, it requires the recognition (and recognition of an intentional violation) of a pattern.  The pattern that you could recognize here is just that “<Something or Someone> walked into a bar” is a very common way to start a joke, and that when you hear that phrase the general expectation is that the following sentence will be something absurd.  This expectation/pattern is violated (intentionally) however, as the following sentence is completely reasonable and mundane.  The ability to recognize that it was intentionally “not funny” in order to violate your expectations makes it funny, hence the term “anti-joke”.  I’ve seen comedians take this general idea even further – intentionally going in front of a live audience and ‘crashing and burning’ in front of other comedians (who are presumably their friends) in order to get a laugh out of the other comedians by intentionally violating their expectations…
• Situational – I don’t know if “situational” is the right word here.  Maybe I mean “observational”.  Who knows!  Anyways – an example of what I mean by this would be:
  • Lets say that one of my friends bends over and his pants rip.  Seeing this, I immediately start singing… “‘Cause winding up with no one is a lot less fun, than a burn from the sun… Or sand in your buuuns!”.  The reason I start singing this song is because him ripping his pants triggered my memory (due to recognition of the similarity…) of the “Ripped Pants” song from Spongebob.  When my friend laughs as a result, it’s presumably because he is able to make the same recognition of similarity…

Anyways, my conclusion in this section here is primarily just that humor (both a person’s ability to make jokes as well as understand jokes) seems to be related in some fashion to their ability to recognize patterns.  I provided five examples above, but its clear to me that there are many more (e.g.: puns, inside jokes).  So now we have three ideas that seem somewhat closely related: Pattern recognition, Intelligence, Humor.  Seems there are already quite a few implications here, one of them being that perhaps that higher intelligence may be correlated with a more “advanced” sense of humor…

[Depression and Mental Illness]

Let me prefix this section by saying that of all the sections in this post, this is likely the one where I am least knowledgeable.  I do have some (relatively brief) first-hand experience with depression, however.  I’m going to over-simplify depression a lot, and focus a lot on this statement from Wikipedia: “Depressed people may be preoccupied with, or ruminate over, thoughts and feelings of worthlessness, inappropriate guilt or regret, helplessness, hopelessness, and self-hatred.”

I wont sugar coat it – there are a lot of depressing thoughts/ideas out there.  These are a few that come to me right now, stream of thought:

• The current state of politics in America.
• The current state of the environment, and the indications that it’s only going to get worse.
• Unless some major scientific breakthroughs start happening, everyone reading this post is going to die eventually.  Beyond that, once you die – there will reach a point where no one will know anything about you and who you once were.
• You’re likely going to have to deal with the death of your parents if you haven’t already.
• We breed animals for the sole purpose of eating them.
• We keep animals in cages for their entire lives for our entertainment (zoos, pet birds, lizards, snakes, fish), taking away their freedom.
• You may never find a fulfilling relationship.
• You may never do anything that you view as truly worthwhile.
• Things like child trafficking, child labor, child sexual abuse, etc. exist.
• Governments and corporations are storing your private communications, and all it takes is one malicious actor out of the bunch to abuse that power and potentially ruin your entire life.
• There are people starving to death as I type this blog post.
• Looking at pay, society seems to value celebrities and professional athletes more than they value scientists and engineers.

I think once you start to ruminate or fixate on any one of the infinitely many depressing topics too much, you start (inevitably) seeing things that remind you of the depressing topic more and more.  You start recognizing patterns, making links.  You start finding new, related things to be depressed about.

Before – you’d start your day grabbing a cup of tea in the lounge at work with the TV on in the background.  With the recent election, however, you got tired of seeing / hearing about Trump so you started avoiding TVs where possible.  As a result, you also started avoiding getting your daily cup of tea.  Now every time you desire tea or hear/see a TV, your mind has linked both of those actions as being related to the negative stimuli that is Trump.  The idea of Trump, and the fact that he’s not going away any time soon becomes more of a fixation for you as a result.  You just want to be sitting on a beach, far far away.  You get phone calls from friends, and they inevitably end up talking about Trump at some point.  Hearing the phone ring now becomes linked with Tea, TV, and Trump.  You stop taking phone calls and end up isolating yourself.  You can’t remember the last time that you were truly happy – it feels like it’s been forever.  You get lonely, and you think about calling your friends – but the anxiety of knowing that they might bring up Trump prevents you from reaching out.  You start avoiding your co-workers for the same reason that you’re avoiding your friends.  Your performance at work slips.  You decide you need some help.

You go to a therapist, and after 3 months of sessions they ultimately end up recommending drugs.  You don’t think anything is wrong with your reaction though, because reality is truly depressing.  You’ve explained this to your therapist ad nauseum, and they seemed to agree – so why is it considered abnormal to react accordingly?  It seems more abnormal to continue to think positively in the face of such depressing things… You stop going to therapy.  You start self-medicating.  You recognize that your life is not anywhere near where you imagined it would be when you were still young and naive.  You have no choice though.  This is Trump’s fault.  This is society’s fault for voting in Trump.  You can’t escape the thought that a single person (in all likelihood) is not going to be able to change the opinions/direction of society as a whole.  Society (and people) as a whole become linked (in your mind) with Tea, TV, Trump, and Phones Ringing.  The interaction required to go grocery shopping becomes unbearable… the pattern recognition continues.

That example was simplified, possibly exaggerated in order to make a point.  It’s clear that pattern recognition and depression are linked, at least in my mind.  If you do a few quick Google searches, it’s clear that Intelligence + Humor are often linked, Intelligence + Depression (or more generally, Mental Illness) are often linked, and Humor + Depression are often linked (see Robin Williams).

My thoughts at this point seem to be – while an elevated ability to recognize patterns may allow you to learn quickly (high intelligence) and have a quick wit – it seems like the same elevated ability to recognize patterns can also have more extreme negative side effects, increasing the likelihood of mental illness.

[Addiction]

This section seems like more of a stretch to me than the others, but I did still feel that it was important to examine the relationship between pattern recognition and addiction.  It seems abundantly clear that addiction is based on a pattern – people repeat an action to the extent that it disrupts their ability to function in society.  Things like smoking, drinking, doing drugs, overeating, gambling, sex addiction, stealing, etc… are all things that people can become addicted to.

These types of patterns, however, seem somewhat different than what I’ve explained in the previous sections.  Above, I was talking about conscious patterns – ones that can easily be verbalized.  In the case of an addiction, it doesn’t seem quite as ‘conscious‘.  I can’t imagine a person who is addicted to smoking being of the conscious mindset “Every day I’ve smoked, something positive has happened.  It seems clear to me that there’s a relationship/pattern – smoking is correlated with positive things.  I need to keep smoking” – but who knows.

It seems like the patterns being recognized in the case of addiction are not being recognized by the conscious mind, but rather by the subconscious.  Some chemical reaction is being triggered, and after some repetition – the body recognizes the pattern and adapts; coming to ‘expect‘ the chemical reaction.  When it doesn’t come, the body is thrown off because the established pattern is broken.

I don’t really have much more to say on this topic – but I am curious if there is a relationship between conscious pattern recognition and subconscious pattern recognition.  If that was the case, it seems like it’s not much of a leap to correlate Intelligence + Addiction, Humor + Addiction, Mental Illness + Addiction, and possibly more…

[Transference of Patterns]

It’s very common for people to pass on advice from generation to generation.  Over a person’s life time and based on their past experiences, they can usually provide you with some fairly generic (but overall fairly accurate) bits of advice to live by.  This information can be communicated through parents, schooling, religion, reading, etc.

I believe that this is how many religions start in the first place.  A collective of people have a seemingly consistent explanation of reality / morality, and they record it.  Their explanation of reality is based off of the patterns that the collective saw in their own lives.  For example – maybe the collective observed that in the overwhelming majority of cases, killing other people was a bad thing.  It brought more sadness than happiness, and it destroyed human potential.  Therefore, the collective in their wisdom came up with the advice “Thou shalt not kill” and included it in their book along with a few stories to drive home their point.  Similarly, once the collective started sharing more stories – it became apparent that many of them had witnessed what they could only describe as miracles.  One member of the collective suggests that maybe there is some form of higher power performing these miracles.  The collective thinks this is reasonable (albeit not falsifiable), so they start building a narrative in their book about this higher power.  The higher power becomes an explanation for all things that are unknown to the collective.

Many religions exist and/or have existed.  Initially, I imagine that their ability to propagate is based on whether or not the ideas presented in the religious texts agree with those of the reader(s).  If the ideas in the religious texts are consistent with what the reader has observed/determined in their own life, they may start adopting the explanations provided by the religious text.  Then after reading and studying the book more, the new convert is starting to see patterns in his day to day life that correspond with the teachings of the religious text.  He may start blindly accepting portions of the religious text as factual, since it seems like so much of the religious text agrees with his version of reality already.  I like to think of it this way – based on life experiences, lets say that this is my running explanation of reality – expressed as a simplified numerical pattern:

2, 3, ?, 7, 12, 13, ??, ??, 23, 28, 31, 37, 41, ??, 47, 53, 59, ??, ??, 71

This is what is presented by the religious text

2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71

Due to the similarities between the religious text and what I had already established based on my life experiences, I adopt the explanations provided by the religious text.  I would say this adoption is possibly motivated (to some extent) by what I’ve been referring to as pattern recognition.  I think religions (regardless of whether or not they’re right or wrong) have served an important purpose of transferring the collective knowledge/observations of society down through the generations.  They provide answers to some questions on morality that no one really has the answers to, and I’d say that most of the widely-propagated religions out there provide answers that are reasonable most of the time.  It’s not until you get into those edge cases involving deciding between killing babies vs. killing nuns that you start to see the rules diverge…

With more modern developments like “deep learning” artificial intelligence, I’m interested in seeing if we start finding more complex methods to store / transfer the general patterns/knowledge/experiences of humans.  Methods that will evolve over time in an automated fashion, but also methods that will then be able to be translated back into a more simple, human representation that we can “digest”.  Maybe there will be a religion based on an AI God.  The AI God will digest massive amounts of human data by scraping the web and interacting with various humans and then based on that – it will spit out some sort of a “Bible” that will put the AI God’s findings into words.

My only “conclusion” in this section is that once a human recognizes a pattern, they translate it internally into some form of general rule.  These “rules” show up all the time – not only in the context of religion, but also just in popular culture.

[“Bubbles” of Thought and Extremism]

With things like 24/7 news, Facebook news feeds, Reddit, Google News, Twitter, etc. there is a constant flow of information being consumed by people.  It’s obvious that no one has the capability to consume all of the information effectively, so they start applying filters.  Perhaps they start only looking at technology-based, privacy-based, and cryptography-based news feeds – because they’re the topics that they find the most interesting.  At this point, you’ve already started forming a “bubble” – as you’re excluding a lot of information from your consumption.

Now perhaps lets say in your technology-based news feeds, you start seeing arguments of which language is better – Python vs. Java.  Based on your coding experience, you have a preference for python.  As you start reading the arguments, you find yourself forming a stronger opinion.  You start subconsciously putting the arguments into generalized buckets, recognizing their similarity – recognizing patterns.  You reach the point where most any pro-Java person could come up to you and spout off an argument in favor of Java, and you can immediately see that it falls into one of the generalized buckets you’ve established – and you say “I’ve already heard that argument.  Try harder.”  Now, when you see news articles that are pro-python you don’t even bother checking the validity of the article, because it fits the pattern you’ve already established in your mind.  Python is king.  This is one way that fake news can be used to divide a population…

It seems to me that with an increased ability to recognize / form patterns, that one would have an increased tendency to form more ‘extreme‘ or ‘unwavering‘ viewpoints, and to sort-of surround themselves with people of similar mindsets.  I’ll provide a lazy example below.

Lets say that this is “true” reality, represented as a sequence of letters:

ABAABABBCABAABABBCCABAABABBCCCABAABABBCCCCABAABABBCCCCC…

Lets say that there’s one group of people – based on their personal observations, that believes that reality is as follows:

ABAABABBCCCABAABABBCCCABAABABBCCCABAABABBCCCABAABABBCCC…

Lets say, additionally, that there’s a second group of people – based on their personal observations, that believes that reality is as follows:

ABABABABCABABABABCCABABABABCCCABABABABCCCCABABABABCCCCC…

Now lets say that group #1 and group #2 are in constant disagreement over who is right and who is wrong.  The people in each group that are best at pattern recognition are the most outspoken, extreme members of the group – because they’re the ones who are able to “best” recognize the patterns and similarity between the proposed explanation of reality, and their own personal reality.

My proposition in this section is that increased ability to recognize patterns = increased likelihood of extremism.

[Auto-Pilot]

In everyone’s day to day life, there are some actions that have been repeated so many times that you’re effectively operating on auto-pilot when you’re performing them.  I know that personally, I’ve experienced the following:

• I’m in the process of showering in the morning, and I realize that I’m not entirely certain what I’ve already done.  Have I already washed using soap?  Have I already done shampoo?  Before the ‘realization’ occurs, it’s clear that I was operating in a sort-of ‘auto-pilot’ fashion.
• I’ll be either on my drive to work, or arrive at work – and realize that I do not recall much at all on the specifics of the drive to work.  I was driving in ‘auto-pilot’ and not recording much mentally.
• I use to arrive at school and put in the combination to my locker in order to retrieve my books.  On most mornings, I recognize that I put in the combination without much conscious thought at all.  In fact, on the days that ‘snapped out’ of the auto-pilot, I found it very difficult to remember what my combination actually was.  I think this is sometimes referred to as just ‘muscle memory‘.

It’s clear in all of these auto-pilot examples, that the person has subconsciously recognized a pattern of sorts – and that the person is able to now perform the actions without much conscious thought.  This seems to have implications in a lot of areas – such as professional sports.  Part of acquiring a skill like being good at soccer is to repeat actions over and over again until you’ve effectively formed that ‘muscle memory’ and you no longer have to put conscious thought into it.

I’m curious if there’s a link between this idea of “pattern recognition” that I’ve been establishing, and the frequency at which a person is operating on “auto-pilot”.  This seems to have somewhat horrible implications if taken to the extreme – ones that would imply that a more “intelligent” person would actually perceive time as moving faster, because they’d be operating on “auto-pilot” more.  To expand on that – the more patterns that the person is able to establish over their lifetime, the faster their perception of time would accelerate, as more and more of their life would be spent on auto-pilot.

Maybe that’s why there’s a general recommendation floating around that says more or less that people should try to experience as many new things as they can.  The reasoning behind this might be to break the established patterns.  To try to force themselves out of auto-pilot, and into a more conscious and “slow” state of mind.

I guess my proposition in this section will be that people who have increased abilities to recognize patterns will find it easier to pick up skills, to form ‘muscle-memory’, and to operate on ‘auto-pilot’…

[Conclusions]

Based on this thought experiment, I’d say that this is my TL;DR of theories / propositions that I’ve come up with:

• Increased pattern recognition implies increased intelligence (as measured by IQs)
• Increased pattern recognition implies a more “advanced” humor
• Increased pattern recognition implies a higher probability of depression & mental illness
• Increased pattern recognition implies a higher probability of addiction
• Increased pattern recognition implies a higher probability of extremism
• Increased pattern recognition implies a higher probability of operating on ‘auto-pilot’
• Recognized patterns are sometimes translated into general rules, and passed on through generations
• Links between intelligence, humor, depression, mental illness, addiction, extremism, perception of time, etc. might be explainable by looking at the person’s ability to recognize and establish patterns

At this point, it’s also clear that I’ve glanced over quite a few areas like:

• Can a person have an increased ability to recognize patterns in one area that does not translate to an increased ability to recognize patterns in another area?  Or is pattern recognition a universally-applicable trait that can be applied to all areas.

This idea seems worth exploring more.  I will do some academic research on this topic and see how much my theories / propositions actually hold water.