Dissociating episodic from semantic access mode by mutual information measures: evidence from aging and Alzheimer's disease

A theory of hippocampal function: New developments

We develop further here the only quantitative theory of the storage of information in the hippocampal episodic memory system and its recall back to the neocortex. The theory is upgraded to account for a revolution in understanding of spatial representations in the primate, including human, hippocampus, that go beyond the place where the individual is located, to the location being viewed in a scene. This is fundamental to much primate episodic memory and navigation: functions supported in humans by pathways that build 'where' spatial view representations by feature combinations in a ventromedial visual cortical stream, separate from those for 'what' object and face information to the inferior temporal visual cortex, and for reward information from the orbitofrontal cortex. Key new computational developments include the capacity of the CA3 attractor network for storing whole charts of space; how the correlations inherent in self-organizing continuous spatial representations impact the storage capacity; how the CA3 network can combine continuous spatial and discrete object and reward representations; the roles of the rewards that reach the hippocampus in the later consolidation into long-term memory in part via cholinergic pathways from the orbitofrontal cortex; and new ways of analysing neocortical information storage using Potts networks.

Déjà vu: A botched memory operation, illegitimate to start with

Rather than a natural product, a computational analysis leads us to characterize déjà vu as a failure of memory retrieval, linked to the activation in neocortex of familiar items from a compositional memory in the absence of hippocampal input, and to a misappropriation by the self of what is of others.

Retrograde amnesia abolishes the self-reference effect in anterograde memory

Is retrograde amnesia associated with an ability to know who we are and imagine what we will be like in the future? To answer this question, we had S.G., a patient with focal retrograde amnesia following hypoxia, two brain-damaged (control) patients with no retrograde memory deficits, and healthy controls judge whether each of a series of trait adjectives was descriptive of their present self, future self, another person, and that person in the future, and later recognize studied traits among distractors. Healthy controls and control patients were more accurate in recognizing self-related compared to other-related traits, a phenomenon known as the self-reference effect (SRE). This held for both present and future self-views. By contrast, no evidence of (present or future) SRE was observed in SG, who concomitantly showed reduced certainty about his personality traits. These findings indicate that retrograde amnesia can weaken the self-schema and preclude its instantiation during self-related processing.

The Capacity for Correlated Semantic Memories in the Cortex

A statistical analysis of semantic memory should reflect the complex, multifactorial structure of the relations among its items. Still, a dominant paradigm in the study of semantic memory has been the idea that the mental representation of concepts is structured along a simple branching tree spanned by superordinate and subordinate categories. We propose a generative model of item representation with correlations that overcomes the limitations of a tree structure. The items are generated through “factors” that represent semantic features or real-world attributes. The correlation between items has its source in the extent to which items share such factors and the strength of such factors: if many factors are balanced, correlations are overall low; whereas if a few factors dominate, they become strong. Our model allows for correlations that are neither trivial nor hierarchical, but may reproduce the general spectrum of correlations present in a dataset of nouns. We find that such correlations reduce the storage capacity of a Potts network to a limited extent, so that the number of concepts that can be stored and retrieved in a large, human-scale cortical network may still be of order 10⁷, as originally estimated without correlations. When this storage capacity is exceeded, however, retrieval fails completely only for balanced factors; above a critical degree of imbalance, a phase transition leads to a regime where the network still extracts considerable information about the cued item, even if not recovering its detailed representation: partial categorization seems to emerge spontaneously as a consequence of the dominance of particular factors, rather than being imposed ad hoc. We argue this to be a relevant model of semantic memory resilience in Tulving’s remember/know paradigms.

Distinctive neuropsychological profiles differentiate patients with functional memory disorder from patients with amnestic-mild cognitive impairment

OBJECTIVES

Patients with functional memory disorder (FMD) report significant memory failures in everyday life. Differentiating these patients from those with memory difficulties due to early stage neurodegenerative conditions is clinically challenging. The current study explored whether distinctive neuropsychological profiles could be established, suitable to differentiate patients with FMD from healthy individuals and those experiencing amnestic mild cognitive impairment (a-MCI).

METHODS

Patients with a clinical diagnosis of FMD were compared with patients with a-MCI, and healthy matched controls on several tests assessing different cognitive functions. Patients with clinically established mood disorders were excluded. Patients with FMD and a-MCI were broadly comparable on the level of their subjective memory complaints as assessed by clinical interview.

RESULTS

The neuropsychological profile of the FMD patients, although they expressed subjective memory and attention concerns during their clinical interview was distinct from patients with a-MCI on tests of memory [semantic fluency, age of acquisition (AoA) analysis of semantic fluency, verbal and non-verbal memory]. FMD patients did not differ significantly from healthy controls, but their scores on the letter fluency and digit cancellation tasks were not significantly different from those of the a-MCI patients indicating a possible sub-threshold deficit on these tasks.

CONCLUSION

Whilst subjective complaints are common within the FMD population, no objective impairment could be detected, even on a sensitive battery of tasks designed to detect subtle deficits caused by an early neurodegenerative brain disease. This study indicates that FMD patients can be successfully differentiated from patients with neurodegenerative memory decline by characterising their neuropsychological profile.

Formal Approaches in Computational Psychoanalysis and the Embodiment Issue

In the last decades in the domain of cognitive sciences benefitted from the integration of contributions coming from different disciplines, such as neurosciences, cognitive neuropsychology, I.T., linguistics and so on. The functional architectures approach, either implemented with hard or soft computation, or with mixed models, has been a relevant conceptual tool that has provided a unifying framework for many research attempts in the field. The advancement of new conceptualizations based on embodied cognition, the research paradigm emerging from the discovery of mirroring systems in the animal and human brain, is questioning this unitary approach. In fact, mirroring systems seems to provide an explanation for human behaviour that cannot be easily reduced to a computational description. Here then author will present and discuss some formal approaches to the psychoanalytic theory, which could favour a better integration of disembodied and embodied cognition. These models are based on a topological implementation of the classical Freudian Conscious/Unconscious distinctions and on the theory of Bi-Logic mental functioning proposed by Matte Blanco.

The Operationalization of the Unconscious

This paper presents a brief review of some formal approaches to the modeling of unconscious phenomena. These models allow for the operationalization of the concept of psychodynamic unconscious towards a possible inclusion in the fields of Cognitive Informatics and Cognitive Computing. In particular this paper presents the conceptualization proposed by Ignacio Matte Blanco of the functioning of conscious and unconscious thinking. In his original view the two ways of thinking are conceived as two distinct logics, “symmetrical” and “asymmetrical” logics. In this study the fundamentals of his concepts are identified and re-elaborated in a more formal way, with the aim of developing an operational and dynamic functional structure that evolves over time and that includes the affective value of objects. On this basis a computational implementation of the conscious–unconscious interaction that employs Learning Classifier Systems (LCS) is put forward.

Unveiling the metric structure of internal representations of space

How are neuronal representations of space organized in the hippocampus? The self-organization of such representations, thought to be driven in the CA3 network by the strong randomizing input from the Dentate Gyrus, appears to run against preserving the topology and even less the exact metric of physical space. We present a way to assess this issue quantitatively, and find that in a simple neural network model of CA3, the average topology is largely preserved, but the local metric is loose, retaining e.g., 10% of the optimal spatial resolution.

Reduced specificity of autobiographical memory and aging: do the executive and feature binding functions of working memory have a role?

Autobiographical memory (AM) is built up from various kinds of knowledge, from general to specific, via generative processes. Aging seems to particularly affect the episodic autobiographical information while preserving information that is more semantic. However, the mechanism of this deficit has not yet been thoroughly tested in relation to working memory. This study is designed to investigate, in a group of 100 subjects, the relationships between age, accessibility to different levels of AM specificity, and two main components of working memory: the central executive and the episodic buffer. We used a new task composed of four embedded verbal autobiographical fluencies (VAF) - from low to highest specificity levels - exploring lifetime periods, general events, specific events, and details, plus tasks exploring free recall of episodic AM and updating, shifting, inhibition, and feature binding in working memory. The results demonstrate that age-related difficulties increase with level of specificity of autobiographical knowledge, i.e., from semantic to episodic aspects. Moreover, regression analyses mainly show that increase in age-related deficit with level of specificity of AM is largely mediated by performance on executive functions (updating and inhibition) and to a lesser extent feature binding in working memory. The results confirm in episodic AM the executive/working memory aging hypothesis, and for the first time highlight the role of episodic buffer in associating the various different details of specific events that elicit the conscious recollection.

Uninformative memories will prevail: the storage of correlated representations and its consequences

Autoassociative networks were proposed in the 80's as simplified models of memory function in the brain, using recurrent connectivity with Hebbian plasticity to store patterns of neural activity that can be later recalled. This type of computation has been suggested to take place in the CA3 region of the hippocampus and at several levels in the cortex. One of the weaknesses of these models is their apparent inability to store correlated patterns of activity. We show, however, that a small and biologically plausible modification in the "learning rule" (associating to each neuron a plasticity threshold that reflects its popularity) enables the network to handle correlations. We study the stability properties of the resulting memories (in terms of their resistance to the damage of neurons or synapses), finding a novel property of autoassociative networks: not all memories are equally robust, and the most informative are also the most sensitive to damage. We relate these results to category-specific effects in semantic memory patients, where concepts related to "non-living things" are usually more resistant to brain damage than those related to "living things," a phenomenon suspected to be rooted in the correlation between representations of concepts in the cortex.