[en] Objectives. A tendency to externalize internal information on reality monitoring tasks has been documented in psychiatric patients with hallucinations. Furthermore, previous studies suggest that factors such as the emotional salience of the material, cognitive effort and meta-cognitive beliefs are important contributory factors in this tendency to externalize internal information on reality monitoring tasks. However, few studies have investigated these aspects in hallucination-prone subjects. Also, these factors have never been examined simultaneously. In the following study we wished to examine the effects of emotional salience, cognitive effort and meta-cognitive beliefs on reality monitoring functioning in hallucination-prone subjects. Design. Between-participants group design. Method. One hundred normal subjects were administered a reality monitoring task. Words were presented by the experimenter. After each word, subjects were asked to say the first word that came to their mind. Words varied in terms of emotional valence and cognitive effort (high cognitive effort for words requiring longer latency times to associate a word and vice versa). Following a delay, words were presented consisting of those already presented by the experimenter or the subject (old) and those never presented before (new). For each word, subjects were required to identify whether the word was old or new. If the word was identified as old, subjects were required to identify the source of the word (subject or experimenter). Subjects also completed a questionnaire assessing meta-cognitive beliefs. Results. Subjects were grouped according to their scores on a revised and elaborated version of the Launay-Slade Hallucinations Scale (LSHS). Those with scores within the top 25% were included in the hallucination-prone group (HP) (N = 25), whereas scores within the lower 25% were included in the non-hallucination-prone group (NHP) (N = 25). Results showed that the HP subjects had significantly more source discrimination errors than NHP subjects for self-generated items. In other words, HP subjects tended to misattribute to the experimenter items that they had produced themselves. This pattern was especially marked with emotionally charged material and with words that required more cognitive effort. In addition, HP subjects scored significantly higher on a scale assessing meta-cognitive beliefs compared with NHP subjects. Finally, scores on a scale assessing meta-cognitive beliefs were positively associated with source discrimination errors. Conclusions. These results suggest that cognitive effort, emotional salience and meta-cognitive beliefs all play a prominent role in the externalizing bias in hallucination-prone subjects. The results also provide evidence for the validity of the idea of a continuity between hallucination-prone subjects and psychotic patients with hallucinations on reality monitoring tasks, including a number of contributing factors in the occurrence of hallucinations.