Almost everyone has seen a face in a cloud. The tendency to find meaningful shapes in random patterns is called pareidolia, from the Greek para, beside, and eidolon, image. The German psychiatrist Karl Kahlbaum used the term in 1866, and it has been a topic of psychology ever since.
Researchers offer several explanations. One is that our visual system is tuned for faces from the first days of life, because noticing a face matters enormously to a baby. Another is that the brain runs two processes at once: one that automatically spots face-like patterns, such as two dots above a line, and another that actively searches for faces when something is ambiguous. Carl Sagan suggested an evolutionary story: spotting a face, or a predator's eyes, in the undergrowth could save your life, so a few false alarms were a fair price.
Interestingly, studies suggest that being good at recognising real faces does not predict who sees faces in clouds. The two skills seem to be separate.
Ambiguous images also invite us to read ourselves into them. Hermann Rorschach used that idea in 1921, when he published his inkblot test. Whether such tests measure anything reliably is debated, and we make no clinical claims, but the basic observation holds: when the picture is vague, the viewer fills in the detail.
That is what makes cloud reading work as a game of reflection. The sky does not know you. What you see in it, and what you feel about what you see, comes from you, which is why two people can look at the same cloud and find a dragon and a dove.
