Thirty-eight people with spider phobias (19 untreated and 19 treated) were exposed to a series of 72 slides of pictures of spiders, weapons, and flowers that were randomly paired with either a shock, a siren, or nothing. A posteriori, untreated participants reported an illusory correlation (IC) between spiders and shock. Although the IC was not significantly smaller in treated as compared to untreated participants, no significant IC could be shown in treated participants. For untreated people enhanced shock expectancies were already present on the first spider trials. The IC was paralleled by heightened skin conductance responses (UCRs) on shocks that were preceded by spider slides compared to shocks preceded by other slides. There were positive correlations between IC, on-line expectancies, and UCRs.