Perfect information is a concept related to game theory where every player in the “game” is fully informed of all prior actions. An example of a game that allows for perfect information is Chess. Every move made right from the start of the game is visible to both players.
Applying this to a market, perfect information is when all market participants are aware of the current market prices, their own utility and own cost functions at any point of time. Complete information would mean that all strategies, cost functions, steps of all players are common knowledge throughout the gameplay. Both far from reality, but this concept has been used to solve different forms of games.
The assumption of rationality of buyers is a big one in the case of imperfect information. A rational buyer would not buy anything of lower economic value in exchange for money. However with imperfect information and in most cases also incomplete information it is hard to assess the value of a purchase say in the stock market. In a way, the decision of a “sell” or “buy” is dependent on the decision made by others as it would affect the price of the stock.
Perfect information for all market participants would still not be enough to assume everyone will behave rationally. Even if all information is available, they might not be credible. The future can change and can render the information unusable. Buyers will operate on a layer of personal biases nevertheless. Not to confuse this with complete information. Knowing all possible market scenarios and moves of market participants , although impossible with current technology, will not lead to rational decisions. Even with perfect and complete information we tend to compensate for that in our own way.