A neurocomputational model of stochastic resonance and aging

Abstract

Stochastic resonance (SR) is fundamental to physical and biological processes. Here, we use a stochastic gain-tuning model to investigate interactions between aging-related increase of endogenous neuronal noise and external input noise in affecting SR. Compared to networks that have optimal system gain parameter of the activation function, networks with attenuated endogenous gain tuning at the system level, simulating aging neurocognitive systems with more intrinsic neuronal noise but less plasticity, continue to exhibit the general SR effect; however, this effect is smaller and requires more external noise. This set of finding suggests that determining the optimal proportion of resonance-inducing external noise as a function of internal-system stochastic gain tuning properties promotes unified theorizing about sensory and cognitive aging at behavioral and neural levels of analysis. r

Electrophysiological entropy in younger adults, older controls and older cognitively declined adults

“…Thus, while previous research has suggested reduced sample entropy in spontaneous EEG in AD subjects when compared with healthy age-matched controls , we were less certain if event-related sample entropy would be sensitive enough to discriminate between older controls and older adults who may be in the very early stages of cognitive decline, or if the cognitive performance differences we anticipated when comparing younger and older adults would map easily onto differences in EEG entropy. Drawing upon the existing research literature and the computational models of Li and colleagues (Li and Lindenberger, 1998;Li et al, 2006) we hypothesised not only a reduction in adaptive system uncertainty, or entropy, in older declined adults relative to older controls, but also a more generalised reduction in entropy with age, that is, in the comparison between younger and older adult groups, particularly in the frontal lobes. Interestingly, we observed differences between older controls and older declined adults in the hypothesised direction, but a mixture of lower entropy and a more differentiated pattern of entropy level across regions in the younger adults when compared with older adults.…”

“…Modelling ageing memory involves modelling its neurological and information processing resource base. Some theoretical models suggest that generic properties of brain function, for example, the level of intra-network variability, may be causally related to the patterns of age-and diseaserelated cognitive decline (Li and Lindenberger, 1998;Li et al, 2006). The idea that entropy may be of general functional significance comes from studies that have found reduced uncertainty associated with a variety of age-related chronic conditions Kaplan et al, 1991;Lipsitz, 2002).…”

“…Theoretical models of ageing cognition suggest that an increase in the level of intranetwork variability may be causally related to the patterns of cognitive decline typically observed in older adults (Li and Lindenberger, 1998;Li et al, 2006). In their computational models, Li and colleagues have demonstrated that as the signal-to-noise ratio of the system decreases, there is greater disruption of engaged performance.…”

Associative and strategic components of episodic memory: A life-span dissociation.

“…Older adults' robust difficulties in rejecting rearranged pairs as lure are consistent with the proposition that older adults rely more on familiarity signals in memory retrieval and show a reduced ability to recollect specific features about past events (Daselaar et al, 2006;Healy et al, 2005;Jacoby & Hay, 1998). According to the neurocomputational theory proposed by Li and colleagues (Li & Lindenberger, 1999;Li et al, 2001Li et al, , 2005Li, von Oertzen, & Lindenberger, 2006), decrements in the distinctiveness of representations due to deficient neuromodulation contribute to older adults' difficulties at all stages of learning and memory, such as initial learning, consolidation, and retrieval from memory (cf. Craik, 1983Craik, , 2006.…”

Dopamine modulates attentional control of auditory perception: DARPP-32 (PPP1R1B) genotype effects on behavior and cortical evoked potentials

“…Neurocomputational studies have also demonstrated that the dopaminergic tuning of the signal-to-noise ratio of information processing can affect the representational distinctiveness of representations in associative memory (Li, Lindenberger, & Sikström, 2001;Li, Naveh-Benjamin, & Lindenberger, 2005), working memory (Li & Sikström, 2002), and perception (Li, von Oertzen, & Lindenberger, 2006). In the present study, the performance of any G carriers of the DARPP32 gene was primarily driven by the perceptual saliency of the syllables when processing competing auditory inputs, regardless of the task relevance of the information.…”