Our first experiment was designed to investigate the influence of speech type (direct vs indirect speech) on source memory (i.e., remembering who said what). Direct speech is perceived as more vivid and perceptually engaging. Because direct speech focuses on what was said (i.e., the literal wording), we hypothesized that perhaps indirect speech focuses on who said what. Results of this experiment showed that speech type only influenced source memory for the gender of the agent (that made a certain statement). After reading statements in indirect speech people were better at remembering whether a male or a female made that statement as compared to direct speech (p = .006). However, they did not perform better when the name of the agent was asked (p = .09). There were no differences between conditions with respect to the level of transportation people experienced while reading the story that included the statements. The exploratory analysis suggested that speech was represented at the level of gender rather than the level of the individual. This a priori hypothesis was tested in Experiment 2. which was, in essence a replication of Experiment 1 with one change to the materials. All the statements uttered by the female characters were now uttered by the male characters and vice versa. This produced a counterbalancing of the links between gender and utterances across Experiments 1 and 2. If the results of Experiment 1 were somehow due to these links, then we would expect a different pattern in Experiment 2. We conducted a power analysis (based on the results of Exp 1) to make sure we obtained a sufficient number of participants for power at a .80 level. The results of this experiment again showed a significant effect of speech type on memory for the gender of an agent (p = .007), with indirect speech leading to more gender-correct answers than direct speech. This replicates the finding of Experiment 1. Unlike in Experiment 1, this time we also found an effect of speech type on transportation (p = .007), with indirect speech leading to higher ratings than direct speech. The effect of speech type on agent specific memory was not significant (p = .06). Experiments 1 and 2 found an effect of speech type at the level of gender but not at the level of the individual. This could be due to the fact that the characters were not individuated in the story. What differentiates the characters is mostly their name. Gender can be inferred from the name. Experiment 3 was designed to investigate whether comprehenders become better at assigning statements to characters when they have richer mental representations of these characters. In Experiment 3, we provide participants with additional information regarding the characters (their appearance, their home state). This should make it easier to differentiate among the characters and to remember who said what.