Journal of the International Neuropsychological Society (2010), 16, 1006–1017. Copyright © INS. Published by Cambridge University Press, 2010. doi:10.1017/S1355617710000536 SYMPOSIUM Memory monitoring failure in confabulation: Evidence from the semantic illusion paradigm IRENE P. KAN , 1 , 2 KAREN F. LAROCQUE , 2 GINETTE LAFLECHE ,2 H. BRANCH COSLETT , 3 and MIEKE VERFAELLIE 2 1 Department of Psychology , Villanova University , Villanova , Pennsylvania 2 Memory Disorders Research Center , VA Boston Healthcare System and Boston University School of Medicine , Boston , Massachusetts 3 Department of Neurology , Hospital at the University of Pennsylvania , Philadelphia , Pennsylvania (Received December 15 , 2009 ; Final Revision April 20 , 2010 ; Accepted April 22 , 2010 ) Abstract Several prominent models of confabulation characterize the syndrome as a failure in controlled aspects of memory retrieval, such as pre-retrieval cue specifi cation and post-retrieval monitoring. These models have been generated primarily in the context of studies of autobiographical memory retrieval. Less research has focused on the existence and mechanisms of semantic confabulation. We examined whether confabulation extends to the semantic domain, and if so, whether it could be understood as a monitoring failure. We focus on post-retrieval monitoring by using a verifi cation task that minimizes cue specifi cation demands. We used the semantic illusion paradigm that elicits erroneous endorsement of misleading statements (e.g., “Two animals of each kind were brought onto the Ark by Moses before the great fl ood”) even in controls, despite their knowing the correct answer (e.g., Noah). Monitoring demands were manipulated by varying semantic overlap between target and foils, ranging from high semantic overlap to unrelated. We found that semantic overlap modulated the magnitude of semantic illusion in all groups. Compared to controls, both confabulators and non-confabulators had greater diffi culty monitoring semantically related foils; however, elevated endorsement of unrelated foils was unique to confabulators. We interpret our fi ndings in the context of a two-process model of post-retrieval monitoring. ( JINS , 2 0 1 0 , 16 , 1006–1017.) Keywords : Anterior communicating artery aneurysm , Memory disorders , Neuropsychology , Cognition , Frontal lobe , Recognition INTRODUCTION to ruptured anterior communicating artery (ACoA) aneu- rysms. Although the precise anatomical locus of confabula- The study of confabulation has a long history (Korsakoff, tion is still unclear, recent research points to the ventromedial 1889/ 1996 ; Talland, 1965 ), and most researchers charac- prefrontal cortex (vmPFC) as a critical region (Fischer, terize the syndrome as a propensity to confuse untrue mem- Alexander, D’Esposito, & Otto, 1995 ; Gilboa & Moscovitch, ories with true memories (Schnider, 2008 ). The condition is 2002 ; Turner, Cipolotti, Yousry, & Shallice, 2008 ). Given the sometimes referred to as “honest lying” because individuals behavioral and lesion characteristics that underlie confabula- who confabulate are often unaware of the inaccuracies of the tion and the association between frontal lobe damage and retrieved information (Gilboa & Moscovitch, 2002 ). In fact, executive impairments, several prominent models have fo- confabulators frequently insist on the veracity of these dis- cused on the contribution of the frontal executive system torted memories and sometimes base their actions upon them and its role in memory retrieval (Burgess & Shallice, 1996 ; (Schnider, Gutbrod, Hess, & Schroth, 1996 ). Dalla Barba, 1993a ; Gilboa, Alain, Stuss, Melo, Miller, & Confabulation has been linked to a variety of etiologies, but Moscovitch, 2006 ; Johnson, Hayes, D’Esposito, & Raye, the focus of the current study is on confabulation secondary 2000 ; Metcalf, Langdon, & Coltheart, 2007 ; Moscovitch & Melo, 1997 ; Schnider, Gutbrod, et al., 1996 ). It is generally Correspondence and reprint requests to: Irene P. Kan, Department of Psychology, Villanova University, 800 Lancaster Avenue, Villanova, agreed that confabulation refl ects a combination of memory Pennsylvania 19085. E-mail: [email protected] disorder and executive defi cits, as evidenced by the fact that 1006 Monitoring failure in confabulation 1007 confabulation is not often found in amnesia secondary to Melo, 1997 ). In a similar model, Burgess and colleagues damage to the medial temporal lobes (Baddeley & Wilson, (Burgess & McNeil, 1999 ; Burgess & Shallice, 1996 ) pro- 1 9 8 8 ; B u r g e s s & S h a l l i c e , 1 9 9 6 ; J o h n s o n e t a l . , 2 0 0 0 ) . posed that failure to properly constrain retrieval specifi ca- Confabulation is most commonly observed in retrieval of tions can lead to retrieval of “generic representations” (i.e., episodic/autobiographical memories. Thus, most of the ex- routine responses that are consistent with blended versions isting models of confabulation were formulated to accom- of similar past events). modate those fi ndings. Although these models differ in Much of the evidence supporting the idea that confabulators important ways, they all point to impairment in some form are impaired at controlled aspects of memory retrieval has of cognitive control or monitoring as the root of confabula- come from studies of episodic retrieval (for reviews, see tion. 1 Whereas some researchers explain confabulation in Gilboa & Moscovitch, 2002 ; Metcalf et al., 2007 ; Schnider, terms of post-retrieval monitoring defi cits (i.e., determining 2008 ), and some researchers have argued that confabulation whether the retrieved information is consistent with task re- involves primarily episodic and autobiographical retrieval quirements, Dalla Barba, 1993a ; Johnson, 1997 ; Schnider, (Dalla Barba, 1993a ). However, if the core impairment in con- von Daniken, & Gutbrod, 1996 ), others describe confabula- fabulation concerns a domain-general cognitive control mech- tion as the emergent property of the disruption of pre- anism that is required for different forms of memory, we should retrieval cue specifi cation (i.e., specifying what needs to be expect to observe confabulation in semantic memory as well. retrieved) and post-retrieval monitoring processes (Burgess & Consistent with this notion, using a cue word retrieval par- Shallice, 1996 ; Gilboa & Moscovitch, 2002 ; Moscovitch & adigm, Moscovitch and Melo ( 1997 ) found that confabula- Melo, 1997 ). tors produced a comparable magnitude of confabulatory For instance, Dalla Barba’s (1993) and Schnider, von responses in semantic and episodic recall. They argued that Daniken and Gutbrod’s (1996) temporality hypotheses high- previous observations of increased prevalence of episodic light impaired temporal monitoring and an inability to sup- compared to semantic confabulation (Dalla Barba, 1993b ; press active, yet irrelevant, memory traces as the critical Fotopoulou et al., 2004 ) might be the result of differential failures in confabulation (Kopelman, 1987 ; Schnider & Ptak, demands on strategic retrieval (see also Fotopoulou et al., 1999 ). In a less constrained model, Johnson and colleagues 2004 ; Kopelman, Ng, & Van Den Brouke, 1997 ). Specifi - ( 2000 ) attribute confabulation to a generalized failure in re- cally, whereas many tests of episodic retrieval require narra- ality/source monitoring. That is, errors are based on confabu- tive responses, tests of semantic retrieval tend to require lators’ inability to accurately identify and monitor the source single word or phrase responses (Zannino, Barban, Caltagirone, of the retrieved memory, and such monitoring failure can be & Carlesimo, 2008 ). When retrieval demands are matched manifested as contextual errors (i.e., placing true memory in an across memory domains by requiring narrative responses in inappropriate spatial, temporal, or event context; Ciaramelli, both cases, the extent of semantic and episodic confabulation Ghetti, Frattarelli, & Ladavas, 2006 ; Conway, 2005 ; Dalla is equated (Gilboa & Moscovitch, 2002 ). Taken together, Barba, Cappelletti, Signorini, & Denes, 1997 ; Fotopoulou, these data suggest that high retrieval demands may be Solms, & Turnbull, 2004 ; Johnson, 1991 ; Schnider & Ptak, critical in eliciting confabulation. 1999 ) or content-based errors (i.e., recalling inappropriate One way to frame the difference between narrative tasks and information, Gilboa et al., 2006 ; Johnson & Reeder, 1997 ). single word or phrase response tasks is in terms of the demands Proponents of another class of models emphasize that on pre-retrieval cue specifi cation. Questions necessitating a while post-retrieval monitoring failure is an important com- narrative response are typically open-ended and require the ponent, confabulation is multi-determined. In their Strategic participant to suffi ciently constrain the retrieval space to gen- Retrieval Account, Moscovitch and colleagues (Gilboa & erate a suitable answer. In contrast, questions requiring a single Moscovitch, 2002 ; Moscovitch & Melo, 1997 ) proposed that response are inherently more constrained, and thus, rely much confabulation is the result of a failure in several controlled less on a participant’s ability to constrain retrieval space. The aspects of memory retrieval, including guiding and con- fact that confabulation in semantic memory occurs more fre- straining the search process, monitoring, evaluating, and ed- quently under conditions of open-ended retrieval may point to iting retrieved memory, and inhibiting irrelevant memories.