Let me first warn you that the CofS currently considers self-auditing to be dangerous. I disagree because this is where the mind's protection is the most effective. However, the gray band of accessibility can become extremely narrow so that most self-auditing is unproductive of results.
The different types of auditing have a significant effect on how wide the accessibility band is. If you're being audited by a safe and knowledgeable person, you are much more willing to dive more deeply into the hidden portions of the mind than you would be if you were auditing by yourself.
Also, making mistakes Vs having successes is a major factor in accessibility. Early on, when you know the least and are most likely to goof things up, you're better off with somebody who knows what they are doing. On the other hand, a bad auditor who screws things up might actually make the PC less accessible in session than he is when he is alone, hence there has occasionally been a problem with people self-auditing (and not doing very well at it) in between professional sessions that should have fixing them up but instead were making a mess of things.
Here is my list of the types of auditing, starting with the one that gives the greatest accessibility (if done right) and working down.
A) Professional Auditing
Here, an auditor, who should be extremely skillful and knowledgeable, audits the PC (PreClear - from habit we even call OTs PCs). The PC is not expected to know very much. The PC is also expected to pay a lot, possibly with some justification since it is very hard to make a good auditor and the PC is doing very little for himself. Its the unnecessary and screwed up actions that cause people to really scream about the prices (who in their right mind would want to pay thousands of dollars for an unnecessary sec check for example). There is rarely any complaint if the right thing is run to a good result.
I personally believe that this is a cop-out on the PCs part and a way to offload responsibility onto the auditor. Although this will give the widest band of accessibility at any given moment, it does not encourage the band to grow wider to the same degree that learning the subject will. Therefor I think professional auditing is best suited to repair actions and startup actions (to get someone moving) and should not be used as the main thrust.
Furthermore, the PC has little idea of what is going on and he can be led into trouble by bad auditing or incorrect C/Sing (Case Supervision, which specifies what is to be audited). This is the only situation (outside of using drugs etc. to force yourself over your head) where somebody might get worse instead of better because he is usually reaching much deeper than normal because he expects the auditor to keep him safe.
B) Co-Auditing
Here the person studies the materials and is paired up with another person who is doing the same. They take turns auditing each other.
In this case, the co-auditors can get away with the sloppiest techniques and still produce fantastic results on each other because the person being audited is well aware of the procedures and purposes and is taking responsibility. The only really critical factors are ARC and a desire to help. The mere presence of somebody else who is willing to listen and give assistance gives a tremendous boost to this factor of accessibility.
I believe that this is the way to go for most people until they reach a point where they can make good progress on solo auditing.
The ideal situation would be to have a professional auditor available to give assistance in case of trouble.
C) Solo Auditing
Here the person goes at it by himself, but with the full formal procedure of auditing. He uses an E-meter and writes things down in a professional manner.
In the CofS, solo auditing is always done with a Case Supervisor who reads the sessions and issues instructions. This ensures that there is still an external communication line and that someone else is there to provide missing knowledge or an external point of observation.
Although the band of accessibility will be narrower than in a co-audit, this has the big advantage that the person, as his own auditor, has an internal communication instead of having to relay everything in words and it also raises the person's responsibility even further.
The same formal procedures will work without a case supervisor if the person has enough understanding of the subject. The sessions should still be written down, because this helps get everything exterior to the person. The band of accessibility will be found to be narrower because there is nobody else there to bail you out if you get into trouble, but if it is already extremely wide, because of a great deal of study and determination, then it can still be adequate to the task.
D) Self auditing
In the final analysis, the only way out is to become a fully self-clearing individual. If you're floating around in a bodiless condition without a book or an E-meter and there's no one around who knows enough to give you a hand, you should still be capable of taking anything apart and setting yourself free.
But you can't self audit in the absence of understanding and without any workable techniques. People have been known to just spin around endlessly just figure figuring and getting nowhere. If people could do this right on an instinctive basis, we would already have gotten out of here. The problem is that we swallowed a whole lot of wrong data early in our existence and it predisposes us to becoming abberated. So the person's instinct is often wrong and leads him into a downward spiral.
Furthermore, the band of accessibility is at its narrowest in this case because the individual has no help. If you are extremely advanced and very knowledgeable, then it doesn't matter so much because the band is incredibly wide anyway, but if you are just beginning and know little and have little perseverance or determination, then the band might be so tiny that little progress can be made.
There is also the research question of developing techniques which are conducive to successful self auditing. Unfortunately, the CofS has leaned away from this. There are some exceptions, like the book "Self Analysis", but generally this is an untapped area.
My own experience was that I was able to successfully self audit the route 1 procedures in "Creation of Human Ability" based on reading the book shortly after I joined the subject. Although this didn't fully turn on real OT abilities, it was a lot of fun and I believe that it set me up for the big OT key out that I had the following year.
According to the Philadelphia Doctorate Course tapes, the OT drills are self auditable if you know the materials well enough, and they can be done immediately on a certain percentage of new people when they walk in the door (these were what was then referred to as Step 1 cases, people who could run step one without preparation). But this research line was pretty much abandoned.
On the other hand, the Dianetic techniques in DMSMH absolutely cannot be self audited. Something like Dianetic repeater technique is almost dangerous even in professional hands, nothing to say of having people go try it by themselves. But that is mostly because it is a poor technique on any basis. The modern Dianetic techniques are smoother and easier, and possibly could be self audited by a very advance case, but such a case would probably be a Dianetic clear already and the technique would actually be in his way and inappropriate to use.
But there is a self auditable technique for running engrams. It consists of alternately spotting something in the incident and something in the room. This can actually let you recover the unconscious period of an operation and bring up the words being spoken and everything. You just need to persist with it long enough. We still have the problem of accessibility. Some cases are just not up to running engrams. But if they are, this trick will work almost as well as the modern Dianetic R3R procedure, with the difference being that it is slower but does not require an auditor or an E-meter.
It is almost criminal not to have a technique like this and teach it to people as soon as they are up to running engrams. What if the person should die? This is one of the few ways that they could run out the death experience successfully.
Furthermore, this one works for Dianetic clears without the liabilities of running R3R (which can get the NOTS aspects of the case all stirred up if too much of it is done after clear).
The technique would be to "Spot an incident underlying (condition)" and then alternately spot something in the incident and something in the room until something happens. Then spot another incident underlying ... etc. until the condition is handled. In general, you should not ask for an "Earlier Similar" incident when doing this on a clear because the "engram chain" mechanism is not fully operative after someone goes clear. A looser question allows for an earlier similar to be run if one is there but doesn't force it. This is actually more of a high powered recall process rather than true Dianetic picture running.