Investigating Unstable Intuitions

I’ve discussed a technique for investigating your subconscious intuitions. A main way you know about intuitions is they like some things and dislike others (and have no opinion on others). Even if they won’t explain themselves in words, they express opinions – that’s a main way that you know they exist. The opinions are often simple, like positive or negative without explaining reasoning. So you can ask questions and consider scenarios. By getting your intuition to give its opinion on many scenarios, you can get a lot of information about the intuition (including finding out some scenarios it has no opinion on which aren’t relevant). You can imagine a scenario, then repeatedly change one factor at a time to see which factors affect your intuitive responses. It’s kind of like deriving the formula for the intuition by feeding in a bunch of different input data and seeing the outputs.

There are difficulties. Sometimes an intuition may be unsure about a scenario. Sometimes you’ll have multiple intuitions and have trouble differentiating them. If you ask yourself 100 questions and get one intuitive answer for each, but each answer comes from any of five intuitions, and you don’t know which intuition gave which answer, then it’d be very difficult to reach a conclusion about any particular intuition. And sometimes you may think you know which intuition an answer came from but be mistaken.

Another major difficulty is that not everything you think is suitable for investigation. The underlying premise of this investigation style is that you have some consistent ideas, so you can gather many data points about the same idea. This doesn’t work when your answers are excuses that contradict each other. If you’re defensive and making up ad hoc answers to try to avoid being bad in some way, the answers may not make much sense when taken together as a data set. Each answer may be an isolated local optima.

In cases like that, there is some kind of deeper stability. E.g. your answers may be superficially contradictory, but all be consistent with you being defensive.

When a bunch of intuitions, ideas or behaviors appear to be local optima – because they contradict or work against each other – they often are all consistent in some way that you’re missing. E.g. they’re all attempts to deny that you have a particular weakness. So the deeper, stable issue is you’re defensive about that weakness. They may contradict each other regarding times, dates, facts and logic, but all follow the theme of denying you have a particular flaw.

Even if someone’s answers seem careless and pointless, with no real theme, then there is actually still a theme. Answering questions thoughtlessly, without regard to logical consistency, and without basing the answers on some deeper principles, themes or goals, is itself a trait that can recur and be stable over many issues. Being thoughtless, careless, or reckless with answers can itself be the deeper stability. This would generally only apply to some topics or categories for someone, not to everything. Pretty much everyone is more reasonable about some issues or topics.

There are other sources of inconsistent output data when investigating your intuitions. They might give different answers to the same questions on different days. Why? Your intuitions can take into account context, like whether you currently feel defensive, upset or happy, even if that isn’t actually part of the question or scenario you’re trying to consider. It could be that you felt defensive about something else, and that affected your answers that day, even though the questions and scenarios weren’t actually eliciting defensive reactions.

In other words, if your mood is unstable, it can cause instability in the outputs from your intuitions. To address this problem, it can help to do investigations in short time periods. Your mood tends to be more stable within a single day than over a year. If you can’t finish an investigation in a day, a week is better than a month. Avoiding break days can also help – sometimes your mood resets when you spend one or more days without working on a project. You can also take a break before starting the investigation in order to try to get a mood reset.

However, if you do the investigation all in one day, then you may find the results are invalid on future days. If you were having a bad day, you may find your intuitions are different on a better day in the future. But at least you got some results about your intuitions on bad days, which is easier to analyze than more chaotic data from a mix of good and bad days.

Changes in moods can happen due to what ideas you encounter and are reminded of, social interactions, successes, failures, and many other events. Your mood may also be affected by your diet, nutrition, sleep, fatigue, etc. These factors may vary from day to day.

If you have poor awareness of your moods, then you may investigate an intuition while in several different moods, get inconsistent answers, and incorrectly conclude that the intuition is illogical when it’s actually just mood-affected.

Learning to be aware of your moods – without trying to control or harshly judge them – is an important step towards having better self-understanding. One of its benefits is helping enable better data collection about your intuitions. It also helps you avoid participating in discussions when you’re tilted, frustrated, upset, agitated, etc. Or you can recognize your mood and participate differently – e.g. acknowledging and exploring what’s going on. Some people have such poor awareness than they can’t recognize that they’re tilted even after a discussion partner recognizes it and tells them. A lot of people seem to habitually deny being tilted, without actually thinking about it, when they’re tilted. People with poor self-awareness about their moods aren’t in a good position to have effective philosophical discussions and have a harder time investigating their intuitions.