Shops and restaurants which play music in a bid to attract customers should heed a warning note from Leicester University: the wrong music can be worse than no music at all.
Although there is widespread agreement that music has a powerful effect on its hearers, there is an extraordinary lack of published research on how it works in real-life situations.
Adrian North, a research student in Leicester's psychology department, has been testing the laboratory-based research theory that people prefer moderate complexity music to high complexity music, such as modern atonal classical works and avant garde jazz, or low complexity sounds such as repetitive rave music.
He set up a music-playing advice stall in the student union cafeteria, playing ten-minute segments of high, low and moderate complexity New Age pop music, and moderate complexity Wurlitzer organ music, interspersed with silence.
The students hated the mighty Wurlitzer, and were also unimpressed by the high and low complexity music, although they liked the moderate complexity New Age bands.
There was a direct link in questionnaire results between how much they liked the music, and how much they said they liked the cafe and said they were likely to return.
They were ambivalent in the questionnaire about how likely they were to visit the advice stall.
But Mr North found that when the students' preferred music of Enya and Enigma was playing, around a dozen people came to the stall, with around five turning up when there was silence, and only around three during the disliked music.
Mr North also played moderate and high complexity New Age music and moderate complexity military music as an accompaniment to a Cadbury's Flake advert, and again found a direct link between how much people enjoyed the music, and how much they liked the advertisement, including its visual impact.
The military music was not a success, showing that the complexity of the music must be linked to how appropriate people find it.
Laboratory-based research has also suggested that people used music to moderate the emotional impact of a situation, while Mr North's research, which included aerobics and yoga classes, showed that people preferred music which heightened emotional impact.
He said: "They're using music as a tool to get more from the situation. It's not just sonic wallpaper there in the background."