Many students think of paraphrasing as what you do when you read a source and then "put it in your own words." This concept is somewhat accurate, but it's not the best way to think about paraphrasing. You should not be trying to find replacement words whose meaning is identical to the language of the source, and then simply put the new words precisely into the place of the original words. That's very hard to do, and often it results in roundabout, lumpy sentences that don't really make a lot of sense. What if you read that "Therapy dogs provide physical and emotional comfort to the elderly in rehabilitation facilities." You would never want to write in your essay, "Healing canines bring rest and repose to old people in hospitals." If you do so, you make your sentences like math equations—plugging in words that you think are equivalent to other words. These are not usually good sentences.
When you paraphrase a source, you are not rewording so much as restating or recasting the idea in your voice, your language, and in a context that assists the purpose of your paper. Here are some examples.
Original source: Jerry Adler. "Thinking Like a Monkey." Smithsonian January 2008: 58-62.
"Santos's interest here is in what psychologists call 'theory of mind,' the ability to impute thought and intentions to another individual, one of the cornerstones of human cognition . . . As recently as a decade ago, the conventional wisdom doubted that even chimpanzees, which are more closely related to humans than are monkeys, possessed theory of mind. This view is changing, in large measure because of Santos and her collaborators . . . The experiment relies upon one of the most predictable traits of rhesus monkeys: their tendency to steal food at every opportunity. Santos discovered this a few years ago when she and her colleagues were running experiments in cognition and tool use involving lemons, and frequently had to quit early because the animals stole all the fruit. The island's monkeys are supplied with food, of course, and they also forage, but to leave so much as a raisin unguarded is to invite larceny; the researchers eat their own lunches inside a locked cage of cyclone fencing. The theory-of-mind experiment is designed to test whether the monkeys, who obsessively guard their own food, assume that people do the same."
Recently, researchers have studied the intelligence of monkeys through the monkeys' ability to understand that other creatures think. Because rhesus monkeys are known to be skillful food thieves, Laurie Santos and her students used the monkeys' bad habit to find out if a monkey tries to imagine what a person is thinking in order to try to steal that person's food. According to the study, some monkeys act as if a human thinks like a monkey.
You see, this passage is not a summary of the article because it does not cover all of the author's main points, such as the methodology of the experiment or its specific results. It doesn't even cover all of the main ideas of one section of the article. The new passage is a paraphrase because it expresses the source's idea—the purpose and description of the thieving monkey experiment—in language that is largely different from the language of the source.