This thesis reviews experimental investigations of psychic readers, and concludes that there is little reason on the basis of this evidence to believe that they have paranormal access to information about their clients. However, a random sample survey of 1,000 residents of Edinburgh district did find support for the claim that the general population is sympathetic to the claims made by psychics. These findings are reconciled with reference to Pseudopsychics' claimed ability to simulate psychic abilities through the use of a technique known as cold reading. A model is proposed, informed by a review of pseudopsychic literature and a pilot study with a known cold reader, which suggests that cold reading actually consists of a number of discrete but interdependent techniques. Central to the model is that much of the reading is dependent on the Barnum effect for success. Experimental work assessed the previously untested assertion that pseudopsychic statements are capable of inducing Barnum acceptance, and found that such terms perform in a similar manner to classical Barnum statements. These statements were used to expand the Barnum pool so that the nature and causes of Barnum acceptance could be studied more systematically. One study explored those properties inherent in Barnum statements which are regarded as contributing to their ready acceptance as true of Ss. It was found that acceptance of items could be predicted on the basis of independent judges' ratings of eight statement properties. Two further studies presented Barnum items as pseudo-feedback from an ostensible psychic reading. These were conducted to explore a proposed model which suggested that Ss accept items because of an artefact of cognitive processing, whereby Barnum statements are not assessed for accuracy in their given form, but rather are interpreted by the client in terms of their own particular circumstances and concerns. Predictions were made on the basis of the artefact model about Ss' recall for the content of the reading, and provided some support for this characterisation of the effect. A final study was conducted to assess the contention that experimental tests of psychic readers misrepresent the function of the reading, and makes the suggestion that with regard to psychic functioning, the client may actually be an active participant. The implications of these results for testing and evaluating psychic readers are discussed.