Verified by 
Most scientists who study dreams use the ‘continuity hypothesis” to explain the content of people’s dreams. This hypothesis states that the content of our dreams reflects our everyday waking experiences. Men are most often the aggressors in dreams because they are the most often the aggressors in real life. Our family members and friends are most often the main characters in our dreams because they are the most frequent characters we interact with on a daily basis and so forth.
Whenever you read a paper on dream content, its results are almost invariably related to the “continuity hypothesis” –as if it were obligatory to do so. While it may be reasonable to adopt the continuity hypothesis as the sort of simplest one out there, or use it as a kind of uninteresting default option on dreams, we are nearing a time when we may have to discard the hypothesis if we want to make progress on understanding dreams.
Why must we discard the continuity hypothesis?
First it is empirically inadequate. It is so easy to come up with very common dream contents that simply do not reflect our everyday waking experiences that I sometimes wonder how it was possible to entertain the hypothesis at all!
Take for example the universal finding that many dreams are filled with bizarre elements (e.g. bizarre otherworldly settings) and still others (e.g. flying dreams) pretty consistently violate the laws of everyday physics. Then there is the fact that many of our everyday life activities (e.g. reading, writing, arithmetic etc) never or only very rarely make it into dreams.
Many of us have sex with a partner on a regular basis but sexual dreams are really quite rare. Half of the characters that appear in dreams are unknown to the dreamer. The dreamer has never met them in real life-yet they consistently comprise half of the characters we interact with in dreams.
Revonsuo and his colleagues have repeatedly pointed out that a great many of our dreams contain simulations of ‘threats’ to the dreamer. Yet most of us do not experience real threats on a daily basis. Revonsuo et al explain this fact by arguing that the threat simulations that occur in dreams are functional. They helped our ancestors to survive in past environments so we moderns are saddled with some threat dreams today. These few examples are just the tip of the iceberg when it comes to the empirical failings of the continuity hypothesis.
Another problem with the continuity hypothesis is that when dream contents are not consistent with the theory, supporters of the theory often offer post hoc explanations to make the data fit the theory. This is not to denigrate such efforts as all scientists do this when they think they have a useful theory. They say “Lets tinker with it a bit before concluding that we need to throw it out altogether”. But sometimes the data that do not fit the theory become so ubiquitous and unwieldy as to strain the capacity of the tinkerers to adjust the theory to fit the data. I think we are at that point now with the continuity theory.
Take for example the well-established finding that content of dreams can affect subsequent life to a considerable amount. Now if dreams merely reflect everyday life then how is it that they can influence everyday life? Supporters of the continuity hypothesis argue that the theory can be adjusted to accommodate this fact but I do not see how.
Must we say that the only experiences reflected in dreams are experiences not previously influenced by the dreams themselves? Any what about those days influenced by dreams? Do we never dream about them?
Most damming for the continuity hypothesis in my opinion is that teaches dream researchers to ignore what is most interesting about dreams--what is most distinctive about dreams. It ignores the fact that dreams have their own logic, not reducible to waking life. The dreaming Mind produces a unique species of cognition. If we do not recognize this elementary fact we will never solve the mystery of dreams.
