Domain-Specific Memory Errors

Expertise in a specific domain can lead to exceptional memory performance, such that experts can typically encode and retrieve large amounts of domain-related information. For example, chess experts can reproduce the exact locations of chess pieces from a game that is in progress (Chase & Simon, 1973; de Groot, 1966), baseball experts can recall large amounts of baseball-related information (Voss, Vesonder, & Spilich, 1980), and a college student used extensive knowledge of track and field to accurately recall long strings of rapidly presented digits (Chase & Ericsson, 1982). However, there are some costs associated with expertise, such that memory accuracy may actually decline in certain situations. For example, although doctors are more accurate than 3rd-year interns at making diagnoses, they are worse at recalling the exact information they used to make a diagnosis (Patel & Groen, 1991; Schmidt & Boshuizen, 1993). Arkes and Freedman (1984) found that on a recognition task, baseball experts made more false alarms to domain-relevant distractors compared with nonexperts, and Baird (2003) found that business majors displayed more domain-relevant intrusions than other students when recalling esoteric business terms. One potential explanation for the benefits and costs of expertise is based on organizational principles of knowledge. Specifically, these benefits and costs may be due to incoming information being incorporated with existing schemata and easily accessed later (e.g., Bédard & Chi, 1992). Van Overschelde, Rawson, Dunlosky, and Hunt (2005) found that participants with high levels of knowledge about American football (i.e., the National Football League) were more likely than lowknowledge participants to recall an ‘‘isolate’’ (i.e., a college football team that was included in a list of professional teams). Although high domain knowledge enhances access to and recall of domain-relevant information, this type of enhanced activation can also lead to increases in memory errors (Roediger & McDermott, 1995; Smith, Ward, Tindell, Sifonis, & Wilkenfeld, 2000). In the present study, we examined whether the organizational processing that benefits experts’ memory performance and leads to a rich encoding context can also increase intrusions during recall. Individuals with high and low levels of knowledge about American football (football experts and nonexperts) studied a list of familiar animal names, all of which were also names of football teams (e.g., lions, broncos, and bears), as well as a control list of body parts. Thus, all of the stimuli were familiar to participants, but the animal names were likely processed differently depending on level of football expertise. We expected football experts to recall more animal names than nonexperts because these names were also names of football teams and thus fit a well-organized schema, but we also expected that activation of this schema would increase intrusions of nonpresented animal names that represented football teams. We expected no differences between groups in performance for the control list.