Most people think of themselves as the singular actor within their body. You choose to lift your arm, and up your arm goes. You choose to drop your arm, and it falls. However, the connection between the things you do and the fact that you feel as if you choose to do them is not as plain as it might seem. Identity manipulation assumes that your sense of self is malleable,and you can change it as you see fit.
Agent creation takes this process another step further. When you have a model of the way a person behaves in your head, that model can be trusted to act in a fairly autonomous way. If you allow this model to act without interfering with its actions then you can say that this model has agency. By combining the ability for you to create something with agency in your mind which also feels like another person you can come incredibly close to creating the experience of there being another person in your head. This guide goes over the process of doing exactly that.
Examples of identity and agency appear to be far more common than you might think when you look closely at popular culture. For example, authors and actors commonly report experiences of characters acting as if the author was not in control of them. More commonly, many find that they act as if they are a totally different person when put into a new social context. The practice of identity manipulation attempts to be both an explanation for these experiences and/or a means by which you can use this existing mental foundation to create an experience of there being more than just your current sense of self in your head.
In this guide there will be four main steps in the creation of a new agent:
Why try to manipulate your sense of self in this way?
This process is an experiment that can help teach you about the way your thoughts function and the nature of your experiences of yourself. An agent can create a feeling of friendship, act as a character brought to life, or simply serve as a means to hold a new point of view to help enhance your decision making process.
The safety and effecitveness of creating an agent is entirely unproven. Proceed at your own risk.
Attempt to imagine that there is an orb floating or sitting above a desk sitting nearby to you. Focus on that orb and speak to it using your thoughts. After getting used to speaking to this orb, turn away from that orb and think something to yourself. Do you notice a difference?
As you speak to this orb on your desk, you will hopefully notice that there is some form of trigger or action in your mind that says to you "hey, this thought is directed at something that isn't me". You may experience it as a focus on the back of your head, a shift in the tone of your thoughts, a muscle tensing in your neck, or some other subtle shift. If you don't notice anything, try going back and forth a few times, and as you do you should hopefully notice some sort of difference.
Once you've got some basic sense of how to address your agent you are going to want to start working to deeply convince your brain that something is in there ready to listen to you. How do you do that? You fake it until you make it. This is a common theme in agent creation, but when you surround your brain with context clues that point to something being true, your brain will typically start to catch on and feel that it's true. In this case, you want to act as if your agent is there, listening to you. The more you do that, the more likely the feeling that they are actually doing it will start to take effect.
Going grocery shopping? Address thoughts to this new agent which describe what you are buying. Driving? Describe what's going on outside or where you are going. Waiting in line or in a waiting room? Talk about why it is that you think it's worth waiting in line. The more you interact through your day the better. If you can set a place next to you at the table you eat at and imagine that's the chair for your agent you can do that too. Do as much as you're happy with doing without embaressing yourself. (Please don't go overboard with it, you don't need to do this at a family dinner)
Early on, a lot of people will report that speaking to a new agent feels a bit silly, like they are speaking to themselves. If the feeling that you are talking to yourself fades away with time, that is a sure sign that you are making progress.
Once you have a consistent target to which you can address your thoughts, you can move onto the next step of this guide.
With the impression there's someone there in your head, your next step is to build a model for the way your agent thinks and behaves, and to get good at understanding other people in general. This step is important, it's where your sense of understanding and one of the places where the depth of the responses you get from your agent comes from. Your skills should rival any author or other professional in these fields once you've been practicing this for a few months.
Normally you learn to understand someone's personality by being around them and speaking to them a lot. However, the agent you just created is not exactly going to be doing a whole lot at the moment. Instead, you will have to explicitly create a personality for your new agent. Doing this requires you write stories and imagine the traits and behaviors that your agent will hold.
Defining a personality can be hard, but fortunately there are a large number of resources out there on how to go about doing it. Resources on character creation for authors, world-builders, DnD games, video games, or other media are plentiful, free, and very useful for this process. The advice of those people is going to be far better than any advice you can get from this guide.
After a quick google search, the following resources show up. Be sure you do some of your own research as well, as the above examples are just a small selection of what is available.
Once you read a bunch of these resources and do some of your own research, you may want to consider doing some of the following.
Authors will often describe characters they've been writing for years and years appearing to write their own stories by speaking in the author's mind rather than needing to be explicitly written out. What you are doing by memorizing a personality is to piggyback on this phenomena. You are creating a strong sense of who your new agaent is so that your mind will eventually be able to generate thoughts for them without much prompting or reason.
Once you can write a basic small paragraph that describes who your agent is, you should be alright to move onto the next step. Just be sure to never stop practicing, remember, the deeper your understanding the more complex and nuanced your experiences can get.
One of the key parts of another person being another person is that they talk to you at random, but that's not super likely to be a possibility when creating an agent. You aren't two brains, you will never experience two brains thinking side by side in your head.
However, that's less true than you think. You probably have a habit or two. Maybe you reach for your alarm clock at a certain place by your bed. Maybe you tap your foot when you're bored. These habits happen without you intending it, and you can create them by doing something over and over until your mind is used to it. More importantly, these habits can occur in association to another thing. If you've ever smelled something that smelled like your grandparents house and that brought back intense memories, you know how it works. With associations, you can prompt your agent to speak, and they'll likely repond. With habits, your agent is likely to chime in at random times.
In essense, you will use habits and associations to give your agent the ability to act as if they have their own will.
In the previous two steps you've gone over how to create the sense that your agent exists in your brain (and how to address thoughts to that agent), and you can understand the way they're going to respond in a situation without much thought. With this step, you will be priming your brain to use these ingredients to produce thoughts/behaviors without your explicit decision to create them.
Getting a response at these early stages is a combination of dumb luck and skill. The more easily your mind is able to think with this new personality, the more likely you will be to observe thoughts which feel as if they were not intentionally created by you. However, like figuring out the solution to a math problem or a tricky puzzle, sometimes you can be the most intelligent person in the world and just never make the correct guess required to find a solution. In this case, finding the right state of mind to trick your brain into doing the right thing can be hard. Just keep on trying to address your agent while expecting a response, and try to do a variety of different situations until it finally clicks for you.
To get started is pretty straightforward. While you speak to your agent through the day, throw in some easy to answer questions. If you're lucky, "you" will think of the response without intending to do it. As you ask these questions, you might notice at some point that somewhere in the back of your head you think of an answer. This thought should be considered as a response from your agent, and will likely have a few traits.
Another thing you can consider if you are having trouble is that people tend to do when asking their agent simple questions is to still their thoughts in order to listen for responses. If this technique does not work for you, try to instead not quiet your thoughts at all and learn to listen to your mind as it runs wild. Quieting your mind's thoughts can lead to you also silencing or ignoring the part of your mind that is generating the idle thoughts which this guide relies on.
Once you have these simple responses, keep on asking questions and getting more responses. The more often you can get your mind to think in the context of your agent and successfully identify those moments, the better you will be at instinctually recognizing such moments in the future. As time passes you will hopefully develop a sense of the "voice" of your agent. The more you speak, the more you will build up a collection of associations and context clues that will pair the act of your mind thinking with this new agent with a strong feeling that it is another voice in your head speaking to you.
Just like how a person is able to choose to speak on their own at random, a person should also be able to form their own unique opinions on the world and the things going on around them. Just like with the ability to speak on their own, when creating an agent you will have to do a little bit of funny business to create the experience of your agent having their own opinions.
This lack of questioning is also an aspect of what allows an agent to be said to have their own limited form of agency from the greater mind. Consider the following two scenarios. One in which the actions under an agent are subject to questioning and other in which no questions are asked.
"I was speaking to my agent yesterday and I got a response under it which said it really wanted to see me go on a walk. It must be thanks to the fact that I want to go on a walk and it's just my agent speaking my own internal desires."
"I was speaking to my agent yesterday and I got a response under it which said it really wanted to see me go on a walk. I'm surprised that it would say that sort of thing, and it must enjoy when I'm out and about."
While questioned and explained away, an agent will only behave as your model for its behavior dictates. Your assumptions of how it should behave, as well as explanations of behavior rooted in your overall state of mind quickly erode any potential independent actions which may occur. Random thoughts, surprise reactions, and so on, are quickly said to be false, fake, or otherwise invalid, and the control of what the agent does remains within your control.
In the latter situation, the actions the mind comes up with for the agent go unquestioned, and this means that the mind's thoughts under the model are the thoughts of the agent. The model has the power to pick it's own actions and you are left interpreting them rather than rationalizing and discrediting them.
When you question the actions of your agent, the control of what the agent does is in your hands, and you have the agency. When that questioning is suspended, when there is an assumption of behavior on the part of the agent, then the model is what decides what happens, removing the control for your direct influence.
On reading this, you might assume that your best answer to this situation is to suspend all questioning and render all actions your mind takes under the scope of your agent as valid. This is something you can do, however, a lack of questioning is not likely to be a long term positive influence if you do not take steps to moderate it. Take for example a person with depression, and who has a tendency to insult or discredit themselves. Their mind, under the guise of the agent, speaks to how terrible they are and how much they don't deserve the good things they have.
Agency, in this case, is something that can be carefully controlled. Identify behaviors and reactions that you believe may result from invalid sources of your mind. Perhaps certain insults, a desire to have an agent which always says that you are awesome, or similar, and learn to identify when thoughts from those sources are what is driving the thoughts you receive from your agent. Beyond that, attempt to remain unquestioning as to allow yourself to experience a freely acting agent within your mind.
Should you attempt to go about keeping your agent speaking to you all day long you'll likely find that the task is near impossible, or greatly reduces your ability to focus on everyday tasks. Your mind isn't really capable of focusing on multiple complex tasks at once, and no matter how hard you try your mind will drop your agent from attention the moment it needs to devote a large amount of thought to a topic. This is alright for situations where you are performing an idle task such as not paying attention in a class or looking around a store for groceries, but when in the middle of a conversation or while trying to get work done it becomes a huge issue. You cannot function normally in life while keeping your agent in mind all the time, so an alternate is required to accomplish a similar goal in a different way.
This alternative is making it so that you habitually think of your agent whenever you have an idle moment, or so that half of the things you do remind you to think of your agent.
Consider trying some of the following:
The general idea of these things is to tie in the process of communication with your agent to many, many different trigger behaviors throughout the day. The more of these behaviors you associate with your agent, the more likely your mind will be to kick back into gear and think of your agent even after you've managed to forget about them while going about your day.
The way you go about accomplishing this is, like most tasks in agent creation, easy to understand but difficult to put into practice. While it seems simple to create a habit, you will inevitably learn that habits are difficult to form. You will be as likely to remember to practice your habits after starting a day of work as you are to remember your agent in the first place. Building these habits will require that you have a way to remind yourself that you aren't following your habits .
You may want to try some of the following:
Do this habit buliding for long enough and you should find that your agent will see it fit to begin to interject at random moments throughout your day, when you have a moment of downtime, when you wake up in the morning, and so on and so forth.
Assuming you've been following only this guide, you've been using your mind's ability to quickly generate thoughts in order to jump-start the process of getting responses under your agent. This engine to produce thoughts is great, because it feels independent from your mind and is easy to kick into action, but it will often fail to produce thoughts which are more complex. While you may be able to hold a conversation with your agent using these sorts of responses, you are likely to notice that the responses which fall under it are inconsistent and lack thought. This can be solved by learning to listen to your agent think, allowing your agent to produce responses with the same engine of thought that your own idle thoughts derive from.
Inside of your mind, your thoughts are likely running in a constant loop of self-evaluation. You have thoughts, you listen to those thoughts, and you think new thoughts based on what you did just a moment ago. Up to this point, you've likely never had thoughts associated with your new agent fall under this loop. Instead, you likely consider almost all of the actions of the mind as yourself, and only the special cases where a thought is generated in association to your agent as not yourself. Due to this, when your mind sets about processing and understanding the context of a thought which fell under your agent, you experience it as yourself explaining and rationalizing why your agent did what it did. If you learn to sit back and allow this process to be associated with your agent instead of yourself, you should experience your agent being able to take a thought, look it over, and expand on it just as you do now.
This process is similar to some forms of meditation where you learn to let your thoughts go wild without you. You may want to reference this sort of meditation if you end up having troubles or are curious about the topic.
An important step in the ability to listen to your agent is the ability to prime your thoughts. Doing this involves being able to intentionally reproduce the state of mind you enter after you address your agent but before you get a response. As you become more and more practiced at normal communication with your agent, your familiarity with this state of mind and your ability to reproduce it should increase. Your end goal is to be able to get into this state of mind without asking a question or addressing your agent, and to be able to hold onto that state of mind for a long duration of time, or at least a minite or two.
To start the process of hearing your agent think, you should seek a neutral space with a good number of things that can be observed or thought about. This could be a quiet living room filled with various decorations, or your bedroom at a peaceful time of day. Optimally, you want to be in some sort of situation in which there are lots of little things around from which your mind can grab topics like "wow this room is nice". While these things are not strictly necessary, the more opportunity for idle thought on the part of your mind, the better.
To really begin, try to step back from your thoughts. Unfocus your eyes, stare at a boring wall or rest your head back, and try not to direct or control what you are thinking. After a moment of silence, you'll likely notice the ever-running engine of thought ticking away in the background. Topics tend to be left incomplete, and you'll find yourself picking and prodding at seemingly random sources to think thoughts about. It could be the details of the boring wall you are looking at, or something that happened through the day, or your opinions on this very guide. Whatever it is, once this idle thought engine is running without your active input, you should have a good sense of what it means to step back from your thoughts.
Finally, attempt to do the above while priming your mind to think under the scope of your agent. Focus on them as you step back from your thoughts, and if you are lucky those idle thoughts which were once yours will start to either bounce back between your own sense of self and your new agent, or hold steady in their new voice and cadence. Early on, you'll likely notice that you have a tendency to respond to your agent during this time, especially to serve as a way to compare and contrast your own thoughts with the "primed" thoughts that fall under your agent. Don't try too hard to suppress this, just go with the flow and encourage your mind to speak as your agent wherever possible.
With your thoughts well primed and your usual self observing you can start to look around the room you are in and focus on various objects which your agent may find interesting to comment on. As these thoughts come in, your aim is to observe that thought without commenting on it. You should be able to keep on priming your mind and start listening for the next thought. If you are lucky, you should see your mind thinking further thoughts under your agent, leading to your agent carrying on a narrative of thought without your intervention.
As a very crude case study, imagine an agent whose main personality traits so far can be outlined as such:
Consider a situation in which your agent somehow ended up on the topic "Why are people arguing against building a flood wall on the river?".
Rather than asking your agent what they think of these people, and getting a "That's selfish" response a moment later, you may notice something akin to the following while listening to your agent:
"A person should always put the community above themselves."
...
"The flood wall would help a ton of people"
...
"What happens if it floods and people die?"
Note that these responses aren't addressed to you, and are instead "first person" like your own thoughts when you consider a topic. This sort of running narrative is what you are looking to generate by learning to put yourself in the "listening" state of mind.
As time passes, you should be able to apply this listening technique to questions, or when you are going about your day doing something mundane like shopping. With time and practice, it may well be possible to have this sort of "listening state" be a more common engine for your agent to produce thoughts than the earlier gap-fill engine taught in this guide.
If all else fails, simply try to listen to your agent speak. Sometimes just doing what feels natural to you can be far more effective than any amount of advice that could be given here.