The effects of hippocampal amnesia on retrieval orientation and novelty processing

Konkel, Alexander

Permalink

https://hdl.handle.net/2142/31117 Copy

Description

Title

The effects of hippocampal amnesia on retrieval orientation and novelty processing

Author(s)

Konkel, Alexander

Issue Date

2012-05-22T00:29:09Z

Director of Research (if dissertation) or Advisor (if thesis)

Cohen, Neal J.

Doctoral Committee Chair(s)

Cohen, Neal J.

Committee Member(s)

• Ross, Brian H.
• Benjamin, Aaron S.
• Gonsalves, Brian D.
• Tranel, Daniel

Department of Study

Psychology

Discipline

Psychology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• memory
• declarative memory
• amnesia
• hippocampus
• novelty
• retrieval orientation
• Goals

Abstract

The medial temporal lobes (MTLs), and in particular the hippocampus, have been the focus of a large body of research since the discovery of a memory deficit in patient H.M. following surgical removal of those brain regions. This research has pointed to the role of the hippocampus in supporting a particular kind of memory, namely relational memory, and has attempted to explain how the hippocampus allows one to retrieve information given some cue (such as on a recognition memory test). However, persistent evidence has also appeared throughout the literature suggesting that the hippocampus is important for novelty processing. Several experiments presented here explore what novelty processing may be exactly, how it relates to typically-discussed memory processing, and how hippocampal function impacts these processes. The broad umbrella of ‘novelty processing’ has been used to describe a few potentially distinct phenomena. One is a differential response to novel as opposed to familiar stimuli, which refers to stimuli with different amounts of previous exposure; a word being seen for the first time in an experiment is novel compared to a word that has already been studied. This will be referred to as stimulus-based novelty. A second is a similar finding but where novel refers to unexpected or unusual stimuli given the experimental context; a green picture presented in a stream of red pictures is novel in this way. This will be referred to as contextual novelty. Finally, novelty processing can refer to the intention to find novelty as opposed to the goal of finding familiar or studied stimuli; the latter ‘familiarity processing’ is what occurs in typical recognition memory experiments. This will be referred to as novelty orientation. Each of these novelty processes has been empirically associated with the hippocampus, although other authors have questioned if the first two actually represent a novelty process per se and the third has received little attention. For that reason, three experiments were designed that investigated novelty as a goal, or more specifically as a retrieval orientation, in contrast to a familiarity-based goal. Previous research has found that participants are easily able to follow familiarity and novelty instructions; as demonstrated by eye movement data they will direct viewing to a familiar stimulus when asked to and will similarly direct viewing away from a familiar stimulus when asked to look at novel stimuli. However, familiar stimuli seem to automatically grab viewing such that it takes longer to direct viewing away. This paradigm (Ryan et al, 2007) was examined using patients with hippocampal damage and intact comparison participants. Famous and non-famous faces were used to provide different levels of stimulus-based novelty, some famous faces were not seen during study to provide a source of unexpectedness or contextual novelty, and viewing instructions were manipulated to provide a contrast between novelty and familiarity orientation. Eye movement data revealed that both amnesic and comparison participants directed different levels of viewing to the different types of faces, supporting the idea that they varied in their memory content. Hippocampal damage did reduce the ability of the patients to distinguish familiar faces from novel lures, but both groups showed the same relative viewing to the studied famous, studied non-famous, and unexpected famous faces; hippocampal damage had no influence on stimulus-based or contextual novelty. Similarly, both groups directed viewing properly following the familiarity and novelty instructions. Recent evidence has suggested that the hippocampus may be important for memory over a short delay in addition to the long-held view of its role in long-term memory. These results were examined along with novelty orientation in a second experiment that expanded on the results of the first. Complex computer-generated stimuli were used instead of faces, and retrieval orientation was manipulated at the trial level instead of across blocks. Behaviorally the amnesic group performed normally when memory was tested after a short unfilled delay, but they were impaired after a long delay. Eye movement data showed that both groups were again able to direct viewing appropriately given the instruction. Critically, there was again no evidence that hippocampal lesions impaired performance under novelty instructions, either behaviorally or with regard to eye movements. Moreover, the eye movement data suggest that participants may have a preference to search for the familiar item at test, raising the question of the necessity of ‘novelty processing’ in addition to familiarity. While other authors have questioned the stimulus-based and contextual novelty effects found in previous research, the first two experiments here implied that novelty processing as a retrieval orientation may also be an unnecessary construct. The final experiment tested that hypothesis more directly using a paradigm that has demonstrated the later consequences of instantiating a particular retrieval strategy or goal. College-age adults studied two lists of words, one under deep (semantic) instructions and the other under shallow (word length) instructions. They then received separate recognition memory tests for these two lists with the knowledge that the old words came from those distinct lists. Previous research (Jacoby et al, 2005a) has shown that participants are able to use this information to focus their memory search, which has the effect of producing deep and shallow processing on the foil items on the two recognition tests; this is demonstrated on a final test where the deep foils are better recognized than shallow foils (called the memory for foils effect). The current experiment manipulated novelty orientation during the middle phase when participants performed the two recognition tests, with one group receiving familiarity instructions and a second receiving novelty instructions. The familiarity group replicated the previous results, demonstrating both a depth of processing effect on the initial recognition tests and a memory for foils effect on the final test. The novelty group produced a reduced depth of processing effect but a normal memory for foils effect. The intention to look for novel stimuli thus does appear to produce different effects than the intention to look for familiar stimuli, although in a complicated manner. The results described here extend the novelty processing literature by attempting to differentiate familiarity and novelty retrieval orientations. Other authors have suggested that novelty processing is not really about novelty at all, but more of a priming effect; that is, brain regions such as the hippocampus are conducting less of their normal memory processing on familiar stimuli than on novel stimuli. The results of the first two experiments support that view; hippocampal lesions had no effect on novelty processing regardless of how novelty was defined. The third experiment instead demonstrated that novelty processing can be distinct from typical memory processing. I discuss why, however, it is unclear that this represents some manner of novelty processing per se as opposed to representing a shift in the meta-mnemonic processes engaged by participants. While the data are unable to definitively determine whether novelty and familiarity orientations are distinct goals or strategies, I demonstrate that novelty processing is not a necessary construct at the mnemonic level.

Graduation Semester

2012-05

Permalink

http://hdl.handle.net/2142/31117

Copyright and License Information

Copyright 2012 Alexander Konkel

Owning Collections