Epilogue: Discussion, Evaluation and Future Research
Harald Maurer in Cognitive Science, 2021
The human neurocognitive system can be regarded as a nonlinear, dynamical and open non-equilibrium system (Glansdorff and Prigogine 1971, Nicolis and Prigogine 1977, von Bertalanffy 1950a, 1953, Schrödinger 1944/2012), which includes a non-equilibrium neurodynamic. In a continuous flow of information processing ("online and realtime computation" [Maass et al. 2002]) the system filters system-relative and system-relevant information from its highly ordered environment, and does so in a manner that integrates new information optimally into the informational structures constructed up to that time ("Free-Energy Principle" [Friston 2010a, Friston and Stephan 2007, Sporns 2011]). Thus, an internal neurocognitive concept consists of a dynamical process which filters out statistical prototypes from the sensory information in terms of coherent and adaptive n-dimensional vector fields. These prototypes serve as a basis for dynamic, probabilistic predictions or probabilistic hypotheses on prospective, new data (see the recently introduced approach of "predictive coding" in neurophilosophy [Clark 2013, Hohwy 2013]; chap. 10.3).
Fundamentals
Arvind Kumar Bansal, Javed Iqbal Khan, S. Kaisar Alam in Introduction to Computational Health Informatics, 2019
Predictive coding is based upon predicting the value of a pixel based on neighboring pixels and past changes. It is a two-step process: 1) predicting the value using neighboring pixels and 2) taking the difference between the actual value and predicted value to calculate the prediction-error. This prediction-error value is small and is encoded. The image is reconstructed using the prediction-error and the predicted value.
Pre-Clinical Approaches and Methods on Alzheimer’s Disease
Atanu Bhattacharjee, Akula Ramakrishna, Magisetty Obulesu in Phytomedicine and Alzheimer’s Disease, 2020
Reisberg (1985), using a global deterioration scale (GDS), predicted that SCD + grade 2 GDS is not normal and grade 3 MCI, grade 4 MCI and higher were associated with dementia. SCD is associated more with depression (Bolla et al.), and depression influences hippocampal volume (Gurvits et al.), suggesting a complex interaction between these three states. The outcome studies have limitation in terms of early dementias, and whether MCI were excluded or not is not clear. Geerlings reported three times greater risk of developing dementia within 3.5 years of being diagnosed as MCI, whereas John and Montgomery and Wang et al. reported that it does not. Cross-sectional studies revealed a greater decline with MCI. Cognitive tests are difficult to quantify between SCD and non-SCD. Biological assessment with quantitative encephalography (qEEG) showed increased θ, whereas voxel-based morphometric studies reported grey matter differences in medial temporal, frontal, and other neocortices. A smaller hippocampus in SCD was reported by van den Filler. Positron emission tomography (PET) studies reported decreased cerebral metabolism of glucose in the parietotemporal and parahippocampal regions. Subjects who had SCD in conjunction with the apolipoprotein E ε4 (APOEε4) allele had greater changes. Higher cortisol levels were reported by Wolf et al. Functional magnetic resonance imaging (fMRI) studies have revealed no change in performance during tasks but observed differences in functional brain activation between SCD and non-SCD subjects. When semantically related words were used, increased activation of the lateral prefrontal cortex was seen and led to compensatory theory, classifying it to be compensation for decay of the hippocampal memory system elsewhere. Testing of the episodic memory and valuation system revealed decreased preference for future-oriented tasks, attenuating attention, and subjective evaluation systems. The phenomenon of “Negative Subsequent” is postulated to be the cause of increased activation of failed recall. Dedifferentiation theory argues that loss of specialization results in diffuse brain activation, but how this hyperactivity contributes to SCD is not clear. Homeostasis breakdown theory says that it could be associated with loss of comprehensive temporal dynamics. Prediction error theory describes brain functions as a statistical optimization engine, and makes implicit predictions of inputs actively making inferences, instead of passively recording. Predictive coding is the principle by which there is a comparison of internally generated predictions with external reality. Awareness of subtle errors can occur in multiple domains, causing SCD. Homeostasis breakdown occurs and compensatory functions act for some time. but these can lead to glutamate excitotoxicity and cell death. Depression is a common characteristic of SCD and there is probably a complex interaction between depression, SCD, MCI, and dementia. Outcome studies are not clear if MCI were excluded. The longer the follow-up, the longer the conversion period.
The response set theory of hypnosis reconsidered: toward an integrative model
Published in American Journal of Clinical Hypnosis, 2023
Steven Jay Lynn, Joseph P. Green, Anoushiravan Zahedi, Clément Apelian
The predictive coding model (PCM; Clark, 2013; Friston, 2010) is recognized as one of the most influential theoretical frameworks in cognitive neuroscience (Beni, 2022). The PCM provides useful insights into hypnotic responsiveness and opens the door to a multifactorial, integrative model of hypnosis. The pillar of PCM is that our brain is not a passive recorder but we act as active scientists, constantly testing hypotheses against evidence (Gregory, 1980). Our cognitive hypotheses derive from heuristic models shaped by prior experiences and will be expressed in predictions propagating downward in the nervous system (Adams, Shipp, & Friston, 2013). The constant comparison between predictions and sensory information shapes prediction errors, which are propagated upward in the system (Adams et al., 2013; Clark, 2013). The free-energy principle assumes that any “self-organizing system that resides at equilibrium with its environment must minimize its free energy” by eliminating prediction errors (Friston, 2010, p. 127).
A celebration of Irving Kirsch
Published in American Journal of Clinical Hypnosis, 2023
Etzel Cardeña
Anticipated by an earlier statement from White (1941), Kirsch’s most important theoretical work has been the response set theory, which posits that expectancies (or expectations when they are conscious) about suggested behaviors and experiences activate responses consistent with them (Kirsch, 1999). In the longest paper in this issue, “The Response Set Theory of Hypnosis Reconsidered: Toward an Integrative Model,” Kirsch’s most frequent collaborator on hypnosis, and an eminent hypnosis author himself, Steven Jay Lynn, along with Joseph P. Green, Anoushiravan Zahedi, and Clément Apelian (2022, this issue) provide an extensive account of Response Set Theory. They describe how it evolved from a “strong” version in which expectancies fully accounted for the variance of hypnotic responses to a more nuanced one in which expectancies are substantial but not the only contributors to hypnotizability (cf. Spanos, Burnley, & Cross, 1993). In what is likely to be a fruitful proposal, they offer an integrative model that considers multiple variables and includes the recent neuroscientific model of predictive coding in which the brain constantly updates a mental model of reality (see also Kirsch, 2018). Predictive coding was clearly anticipated in the non-brain-centric model of constructivists such as Brunner (e.g., “The organism in perception is in one way or another in a state of expectancy about the environment,” Bruner & Postman, 1949, p. 206), and by neuroscientists (e.g., experience is “an inextricable amalgam of represented anticipation and represented perturbation,” Kinsbourne, 1998, p. 241).
Audiovisual speech segmentation in post-stroke aphasia: a pilot study
Published in Topics in Stroke Rehabilitation, 2019
Anahita Basirat, Étienne Allart, Angèle Brunellière, Yves Martin
The contribution of visual cues to speech segmentation could also be explained in the framework of a predictive coding mechanism.45 It has been suggested that seeing gestures while listening to continuous speech allows word onsets to be predicted.46 This anticipation facilitates the breaking up of sentences into words. Importantly, impairment of predictive processing during speech perception has been reported in stroke patients with auditory comprehension deficits47 and in patients with non-fluent primary progressive aphasia.48 The ability of PWA to predict upcoming words while listening to sentences is not known. It would be interesting to assess whether the use of visual speech cues is determined by the presence of intact predictive mechanisms after stroke.
Related Knowledge Centers
- Bayesian Inference
- Confirmation Bias
- Mental Model
- Perception
- Brain
- Neuroscience
- Sense
- Hypercomplex Cell
- Evolutionary Psychology
- Prejudice