Retrieval, reconstruction, and response bias: A cognitive neuroscience investigation of false memories

Our conscious experience of memory often includes reconstuctions of the past in order to fill the gaps of forgotten events and details. The goal of this thesis is to use the tools of cognitive neuroscience to better understand the cognitive and neural underpinnings of such "false" memories. This includes the development of a paradigm and three behavioral experiments demonstrating how false memories can be created in a variety of laboratory settings with high frequency and vividness. When confronted with a complex visual scene, subjects tend to encode only the details of a scene which capture their attention. Details which are not explicitly encoded are reconstructed based on the schema of the scene resulting in false memories (Chapter 2). However, not all errors on memory tests are false memories per se. Recently, a word-list task has been advanced as a paradigm for the creation of false memories. Some nonpresented words are falsely "recognized" as often as studied words are correctly recognized, leading to the assumption that this is due to a false memory for the nonpresented words. However, subjects could also be biased to "recognize" a particular item for many reasons that have little to do with memory. Four experiments were run and analyzed using signal detection methods that demonstrate that the false recognition effects previously reported using this word-list paradigm can be accounted for by shifts in response bias, not false memories (Chapter 3). Conversely, the effects produced by the schematic scene paradigm can be attributed to reconstructive memory processes, not shifts in response bias. Four further experiments show that this reconstruction occurs during the retrieval stage of memory processing (Chapter 4). Finally, a study of patients with unilateral, focal lesions in the prefrontal cortex demonstrates significantly lower discrimination of previously presented items compared to normal subjects. The prefrontal cortex is an area of the brain previously thought to be responsible for processes subserving the retrieval of episodic memory (Chapter 5). Taken together these experiments suggest that our brains seem to compensate for retrieval deficits by either biasing our responses or reconstructing the past.