Well, it depends.
In my experience, I have been to two good therapists (out of two), and they've been pretty perceptive at drilling through the mess and being able to discern patterns.
I know that I was "helping" them, not trying to hinder them, and it was long-term one-on-one therapy.
Someone mentioned the experiments done in 1973, where researcher David Rosenhan wanted to test this concept. So he made up a symptom that could be read as a schizophrenic hallucination (hearing a few "words" or noises, but no other symptoms), and then he and some compatriots tried to get themselves admitted to different psychiatric wards around the country.
They were successful in fooling the shrinks and spent an average of almost three weeks in a ward before being released. Interestingly, about a third of the patients locked up with them and queried about them identified these peopel as "sane" -- whereas the shrinks failed to recognize their sanity. Sometimes even innocuous behavior was written up in the case files as signs of psychological imbalance.
Obviously this paper (printed in Science magazine) pissed off a lot of shrinks. So Rosenhan offered a follow-up match: He warned a specific hospital that he would send a number of patients to them over the next three months, to try to get admitted, and their job was to properly identify them. They complied. in the next three months, the hospital identified (out of close to 200 patients) around 40 to be imposters and another 40 to be suspected of being imposters..
As you might have surmised, Rosenhan had laid a nice trap: He sent no one at all to the hospital. So everyone viewed as an imposter had an actual potential psychological disorder.
Writer Lauren Slater repeated this experiment within the last ten years sometime. If her text is 100% accurate, she was able to check herself into the emergency room and be "treated" and -- while she was not kept for observation -- she was in many cases prescribed medication to alleviate the supposed hallucinations.
Note that Rosenhan's experiment occurred over 30 years ago, and the face of psychiatry has changed since people were routinely locked up in wards. Even Slater's experiment shows more the difficulty that modern-day ER practitioners have in diagnosing people quickly and efficiently while trying to process a large number of people for a myriad of causes:
- Patients commonly expect a quick fix (and thus are prescribed meds of some sort).
- There is a desire to avoid liability -- it's better to treat for a potential cause and have nothing bad happen, than to not treat at all a potential cause and have someone do something horrible.
- The doctor assumes that the patient in fact is working WITH him and not against him -- that he is not lying about his symptoms. There's a level of trust there not just with the patient for the doctor, but the doctor for the patient -- that the patient wants real help and is telling the truth, as much as possible. A diagnosis is only as good as the data it is made with.
To veer off this topic a bit into a different area, such as criminal insanity: In this case, there is an assumption that a criminal might be pretending to be insane in order to avoid punishment. So the psychiatrist is testing the inmate for consistency. While we've all heard horror stories about people "pretending to be crazy," any psychiatrist/psychologist worth their salt and who is healthily skeptical can detect the majority of these cases, where the convict's behavior and story does not match up. Especially, the more experience a shrink has, the better they will be at detecting inconsistencies -- they'll know what to look for.
Any shrink that is easily fooled is either:
- (1) rushing for some reason to a diagnosis,
- (2) too accepting of people's stories at face value, or
- (3) not well educated/experienced.