حالات ذهنی در ارتباط با شکست بازیابی در بیماری پارکینسون

Instances in which we cannot retrieve information immediately but know that the information might be retrieved later are subjective states that accompany retrieval failure. These are expressed in feeling-of-knowing (FOK) and Tip-of-the-tongue (TOT) experiences. In Experiment 1, participants with Parkinson’s disease (PD) and older adult controls were given general questions and asked to report when they experienced a TOT state and to give related information about the missing word. The PD group experienced similar levels of TOTs but provided less correct peripheral information related to the target when in a TOT state. In Experiment 2, participants were given a Semantic (general knowledge questions) and an Episodic (word pairs) FOK task. PD patients failed to accurately predict their future memory performance (FOK) in response to both episodic and semantic cues. Results are interpreted in the context of recent frameworks of memory and metacognition.

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by motor symptoms, caused by a loss of dopaminergic neurons in the basal ganglia. Studies investigating memory in PD have compared two main systems: the episodic memory system and the semantic memory system (Tulving, 1985). Episodic retrieval is characterized by a recollective experience and a feeling of the self in the past (Tulving, 1985). Semantic retrieval is associated with noetic awareness (familiarity and knowing) and reflects conceptual knowledge. Studies in Parkinson’s disease have revealed impairments in episodic memory, particularly on free recall tasks (Daum et al., 1995, Gabrieli et al., 1996, Johnson et al., 2005 and Whittington et al., 2006; see Dujardin and Laurent 2003 for a review). Similarly, many studies have showed semantic memory deficits in Parkinson’s disease particularly on fluency tasks (see Henry and Crawford 2003 for a review). Whilst many researchers have suggested that memory problems in PD are associated with difficulties retrieving information, not many studies have actually explored the subjective states associated with retrieval in this clinical population. Subjective states differ according to the type of memory task involved (episodic versus semantic) as well as the retrieval outcome. Subjective states can either accompany retrieval success or emerge following a retrieval failure. In the case of a retrieval success, subjective states might be captured by the type of contextual information that participants can recall, such as where or when the information was presented (i.e. recollection). On the other hand, when people fail to retrieve the information they might infer that they still might retrieve the answer. Instances in which we cannot retrieve information immediately but know that the information will or might be retrieved later are the subjective states that accompany retrieval failure. These subjective states can be expressed in feeling-of-knowing (FOK, Hart, 1965) and Tip-of-the-tongue (TOT, Brown & McNeill, 1966) experiences. Most studies in Parkinson’s disease have assessed subjective states associated with retrieval success and suggest that impairments vary according to the severity of the disease (Edelstyn et al., 2007, Edelstyn et al., 2010 and Hay et al., 2002; but see Weiermann, Stephan, Kaelin-Lang, & Meier, 2010). However, studies measuring subjective states of retrieval failure in Parkinson’s disease are scarce. The novelty of this study is to explore these subjective states of retrieval failure in Parkinson’s disease by combining both the TOT and the FOK paradigm in two experiments. A Tip-of-the-tongue (TOT) state is the feeling that a currently inaccessible item will later be recalled (Brown, 2012 and Brown and McNeill, 1966). From a linguistic perspective, TOTs signal a partial retrieval failure where semantic information is available, but participants fail to phonologically retrieve the word. TOTs can also be considered from a metacognitive perspective ( Bacon, Schwartz, Paireficout, & Izaute, 2007). Metacognition refers to higher level cognitive processes consisting of a reflection upon the cognitive level ( Nelson & Narens, 1990). From a metacognitive perspective, TOTs signal a partial retrieval success; the partial information retrieved leads participants to experience a TOT. In other words, TOTs are subjective experiences that allow individuals to monitor their retrieval processes ( Schwartz, 2002). An inaugural study published by Matison, Mayeux, Rosen, and Fahn (1982) showed that PD patients experienced word-finding difficulties, thus suggesting that patients were demonstrating some TOT-like states. However, in this study TOT states were not directly measured and patients were not asked to report when they were experiencing a TOT. In a more recent study, Oh-Lee, Szymkowicz, Smith, and Otani (2012) directly measured TOT states using 30 general knowledge questions. Participants could give three different answers for each question: the answer they thought was correct, a ‘do not know’ answer and a TOT answer. In the case of a TOT answer, participants were asked to report the strength of their TOT experience on a scale of 1–20. Participants with PD were found to experience the same number of TOT states as the older adults controls. TOT accuracy was measured using a Gamma correlation comparing the TOT states (using the scale) and the memory performance ( Nelson, 1984). The gamma correlation is an index of the rank order agreement between the metacognitive judgements (e.g. TOT or FOK) and memory performance ( Nelson, 1984). The index is based on the difference between the concordant (e.g. I think I know the word (strong TOT) and I later recognize it) and discordant (I think I know the word (strong TOT) and I do not recognize it) pairs. No significant group differences were observed. Feeling-of-knowing judgments (FOK, Hart, 1965) are predictions about the likelihood of subsequent recognition of currently non-recalled information (Hart, 1965 and Nelson and Narens, 1990). In a FOK experiment participants are presented either with new information to learn such as word pairs (episodic memory task) or are presented with general knowledge questions such as ‘what is the capital of France?’ (semantic memory task). When presented with the question or the first word of the pair they are then asked to recall the corresponding information. If they cannot recall the information, they are asked to predict whether or not they will be able to recognize the missing information if it is presented to them later. This prediction is a FOK judgment. Thus, FOK judgments measure participants’ ability to estimate the likelihood that they will recognize a piece of information they have failed to recall either from semantic memory (e.g. Nelson & Narens, 1990), or from episodic memory (e.g. Schacter, 1983 and Souchay et al., 2000). The most notable difference between the Episodic FOK and Semantic FOK tasks is that the former involves a study phase, so that FOK judgments are made on recently encountered material. On the other hand, Semantic FOKs are made about information assumed to be already stored in memory. A consistent finding is that Parkinson’s patients give inaccurate FOKs on episodic tasks (Baran et al., 2009, Souchay et al., 2006 and Yu et al., 2010). For instance, Souchay et al. (2006) gave participants a list of 20 word pairs to learn and during recall asked them to give a Yes/No FOK judgment when presented with the cue. The results showed that participants with PD were inaccurate at predicting their memory performance. On the other hand, only one study has yet looked at FOK judgments in PD using a semantic memory task (general knowledge questions). Coutler (1989) compared depressed and non-depressed PD groups, finding that non-depressed participants with PD gave accurate FOKs. However, we interpret Coutler’s findings with caution since they were published as an abstract and the methodological details of the exact paradigm used were not provided. These studies suggest that patients with PD demonstrate accurate subjective states associated with retrieval failure (TOT and FOK) when assessed on a semantic memory task (Coutler, 1989 and Oh-Lee et al., 2012) but that they are inaccurate when assessed on an episodic task (FOK studies, Baran et al., 2009, Souchay et al., 2006 and Yu et al., 2010). However, the neuropathological findings in PD support the prediction that all subjective states associated with retrieval failure might be impaired in this clinical population. Most reviews have focused on the dysfunction of prefrontal cortex as a predominant marker of the disease (e.g. Dujardin and Laurent, 2003, McPherson and Cummings, 1996, Owen, 2004, Prull et al., 2000, Taylor et al., 1990 and Troster and Fields, 1995). Dysfunction of prefrontal cortex is the consequence of a reduction in dopaminergic innervation of the basal ganglia and the frontal lobes (see Dujardin and Laurent, 2003 and Lang and Lozano, 1998). Volumetric MRI studies in PD have also documented atrophy of the hippocampus in PD, even in patients without dementia (Bruck et al., 2004 and Tam et al., 2005). To summarize, studies in Parkinson’s disease have shown decline of both the frontal lobes and the temporal lobes in Parkinson’s disease. These correspond to the brain regions highly involved in TOT states or FOK judgments. For example, a set of fronto-temporal and anterior cingulated brain regions are activated during FOKs (Kikyo and Miyashita, 2004, Kikyo et al., 2002, Maril et al., 2003, Reggev et al., 2011 and Schnyer et al., 2005). Similarly, fMRI studies exploring TOT states have revealed a crucial role of the frontal lobes. Maril, Wagner and Schacter (2001) (and Maril et al., 2005) showed that compared to Know or Do not know responses, TOT responses were associated with a greater activation of the right prefrontal cortex. More recently, Reggev et al. (2011) explored the neural correlates of both Episodic and Semantic FOK using fMRI in healthy controls. This study revealed both common and distinct neural correlates. The common regions activated included: the left anterior prefrontal cortex (aPFC), the left middle frontal gyrus (MFG), the bilateral dorsomedial prefrontal area (DMPFC), the anterior cingulated cortex (ACC) and the left lateral superior parietal lobe (SPL). However, the right inferior frontal cortex (RIFC) was more activated in response to semantic cues, whist the left MTG, cuneus and posterior cingulate were more activated in response the episodic cues. In particular, a posterior midline network was activated during episodic and not semantic FOK. Taken together, these findings lead us to predict that participants with Parkinson’s might show impaired subjective states associated with retrieval failure. Contradictory results have been reported in the literature regarding subjective states in Parkinson’s disease, with some studies finding impairments (Baran et al., 2009, Souchay et al., 2006 and Yu et al., 2010) and others not (Coutler, 1989 and Oh-Lee et al., 2012). To explore this issue further, we measured subjective states of retrieval failure in Parkinson’s disease in two experiments; the first assessing TOT states (Experiment 1) and the second investigating FOK states in response to semantic cues (Semantic FOK, Experiment 2) and FOK states in response to episodic cues (Episodic FOK, Experiment 2).

To conclude, the current paper showed that subjective states arising from retrieval failure are not always accurate or of good quality in Parkinson’s disease. Considering the impact that these subjective states have on memory strategies, future research will need to explore how failing to experience accurate subjective states might impact on memory performance. For example, TOT states guide retrieval processes and in particular the strategies put in place, such as time spent to try and retrieve the missing word (Brown, 2012). A difficulty in initiating memory strategies has often been reported in PD (Gabrieli et al., 1996, Knoke et al., 1998 and Souchay et al., 2002) and could be explained by inaccurate subjective states. Finally, future research will need to determine if the accuracy of these subjective states can be improved. In this context, the impact of the dopaminergic treatment should be assessed. Furthermore, recent work from our lab showed that Episodic FOK accuracy in Parkinson’s disease could be improved when memory strategies such as the Self-Performance-Task (SPT) effect are used at encoding to improve memory performance (Smith, Souchay, & Conway, 2010). The next step would be to examine the role of strategic encoding in improving subjective states in PD.